The DOL reported Initial weekly unemployment claims for the week ending on October 6th, 2012 were 339,000, a 30,000 drop from the previous week of 369,000. From pundits to politicians, people are freaking out for this makes initial claims the lowest level since February 16th, 2008.

Ok, we have repeatedly warned, ad nauseum, do not bank on initial claims, especially on the number reported in the press release for that week. Initial claims for unemployment benefits is a weekly statistic and that implies a very short time window for data collection, it is always revised the next week, almost always upward. One can have unusual events which throw off the seasonal adjustment algorithm. One can have missed timing of events that can also throw a monkey wrench in one week's worth of data.

Ok, now everybody notice that October 6th is the 1st week in the third quarter. That means states must do some bookkeeping per the federal law requirements that were part of the various emergency and extended additional unemployment benefits passed by Congress starting in 2008 with the stimulus bill.

States do this normally at the end of the quarter. What happens is States go through the officially unemployed wage records at the end of quarter for those receiving extended or emergency unemployment benefits. It's kind of an audit, required by these various extended and emergency unemployment benefit programs passed by Congress, starting in 2008. If a person receiving benefits from these programs made enough wages during that quarter, they must be reclassified for eligibility under the regular unemployment claims and unfortunately for the eligibility benefit amount to be received, that those new hours worked will determine the new benefit level. That means, even though that person has been receiving extended or emergency benefits, they must refile a new initial unemployment claim and be put into the regular program for the benefit those new wages determined.

Now this number of initial claims filed quarterly is due to federal law and resulting state audit quarterly bookkeeping and normally results in an increase of initial claims. Guess what happened? An unidentified very large state (ahem, cough, California or even New York), reported a decrease instead of the expected increase and didn't get their numbers in on time to boot. States which reported an increase in initial claims number 47.

Take that 47 number to the bank for increases and throw that large state to the wolves for screwing up this week's report, although we won't be able to verify the rogue nation until next week, when the individual state statistics are revealed.

Additionally, there is a surge in initial unemployment claims at the start of every quarter. We can actually see this by looking at the not seasonally adjusted initial claims. The quarterly surge is not just due to the bookkeeping requirements mentioned above, but also due to the benefits computation game for those applying for UI. If one waits to apply for unemployment benefits on a new quarter, that means the benefit amount will be higher than if one applies at the end of a quarter. Unemployment benefit is based on wages earned of the last quarter worked before applying for benefits. Usually people earn the most money right before losing their jobs (nice huh). That means if one waits until the start of the new quarter to file for unemployment, odds are they will get slighly more in unemployment benefits by waiting. The end game result of this fact is we see a surge in new claims at the start of each quarter. In other words, if a person waits to file until Q3, that means their benefits will be based on wages earned in Q2, not Q1.

States missing reporting deadlines, sending incomplete data, general statistical data lag and statistics getting caught on the weekly reporting boundary are more common than the press would lead you to believe. Below is the weekly percentage change of initial claims for the last two years using the seasonally adjusted data. Here you can see the weekly reports are really volatile and not something to bank on. We also see this week's percentage change isn't that unusual.

Last week's initial claims were revised upward by 2,000 after the initially reported 367,000.

Below is the mathematical log of initial weekly unemployment claims. A log helps remove some statistical noise, it's kind of an averaging and gives a better sense of a pattern. As we can see, we have a step rise during the height of the recession, but then a leveling, then a very slow decline, or fat tail. That fat tail is our never ending labor market, stuck in the mud, malignant malaise.

Below is the four week moving average, set to a log scale, from April 1st, 2007. The four week moving average dropped 11,500 to 364,000. Now due to this week's weirdness, that's probably too low, yet we can say for sure at least initial claims didn't ramp up. In the below graph we can see we still are not at pre-recession initial weekly unemployment claims levels. If anyone recalls, even before the Great Recession the job market was not so hot.

Continuing unemployment claims didn't budge, and we have large long term unemployed. The below figure doesn't include those receiving extended and emergency unemployment benefits.

The advance seasonally adjusted insured unemployment rate was 2.6 percent for the week ending September 29, unchanged from the prior week's unrevised rate. The advance number for seasonally adjusted insured unemployment during the week ending September 29 was 3,273,000, a decrease of 15,000 from the preceding week's revised level of 3,288,000. The 4-week moving average was 3,279,250, a decrease of 7,750 from the preceding week's revised average of 3,287,000.

In the week ending September 22, not seasonally adjusted, the official number of people obtaining some sort of unemployment insurance benefit was 5,044,649. More people moved into emergency and extended benefits but these figures aren't seasonally adjusted, so it is difficult to compare. Officially, there are 12.1 million unemployed.
.