...but of course, that drop is because the prior week was adjusted UP to 402,000 (!)

3-week rolling average of initial jobless claims is 374,000.

All the analysts I'm watching are saying it is good news, but the seasonal adjustments and retroactive adjustments like the one from last week are rendering it "less than useful" in figuring out what is going on.

Continuing claims are down to 3.432 million as unemployment benefits run out for many.

U6 -- the more relevant measure of unemployed and underemployed -- has now been reported at 15.2% for December 2011, down from 15.6% in November. That's good news, but as discussed, may be seasonal.

tl;dr : Numbers are getting adjusted wildly due to seasonality and the holiday season. As discussed before, when late Feb 2012 rolls around we'll see where things really stand.

Consumer Price Index up 1/10th of a percent (meh).

Housing starts are DOWN 4.1% in december, vs UP 9.1% in november (bad).

 

-----signature-----

215,000 unemployed Americans fell off the rolls and so are not counted on the U3 or U6, despite the fact that they are STILL UNEMPLOYED

 

-----signature-----

Yeah, for once I'm acknowledging your point (clear... succinct... up front... valid). I'm looking at that too, B_T.

I don't hear anyone talking about that or how that 'magic' works. Unless it has to do with the death rate / retirement rate.

That's probably part of the problem. I think once you hit a certain age you fall off of U6. And we have a LOT of aging boomers swelling the ranks of 'not working but probably need to be' since they lost an assload of retirement money in the recent crash.

I think saw a chart somewhere (not BLS) of what % of the TOTAL population is actually working, effectively supporting the rest of the country (children, elderly, infirm, etc). And that number was scary. I posted it here, once. I may add that to the mix in future if I can find it again.

 

-----signature-----

BLS Definition - "Labor Force"

Labor force measures are based on the civilian noninstitutional population 16 years old and over. Excluded are persons under 16 years of age, all persons confined to institutions such as nursing homes and prisons, and persons on active duty in the Armed Forces. The labor force is made up of the employed and the unemployed. The remainder—those who have no job and are not looking for one—are counted as "not in the labor force." Many who are not in the labor force are going to school or are retired. Family responsibilities keep others out of the labor force.


Lots of room for finagling, with that bolded part. I've seen analysts the last few months start commenting that these numbers... even U6... don't seem as useful as they used to be. A working theory is that we have millions of underfunded Boomers "retiring" who should actually be working. That artificially reduces the current U6, and hides an unfunded liability of picking them up off the street (literally) 10 years down the road when they are out of money and there's no way they can get a job in their 70's.

edit -- I've also seen reports the last month about how as troops return from Iraq, that will swell the unemployment ranks as well.

 

-----signature-----

The number of people on unemployment insurance fell by 215,000.


The unemployment statistics, U3 and U6 included, do not use unemployment insurance stats when calculating the unemployment rate. All of the employment/unemployment data is compiled using surveys of either households or businesses.


The surveys certainly have their problems, but this complaint really isnt valid.

Why is it not valid?

215,000 who are required to be looking for jobs in order to receive benefits as recently as this last month (therefore, still in the market) don't just vaporize. If the surveys are working, that should be reflected as a rise in U6. Or at LEAST a level number. Not a drop.

 

-----signature-----

They still should be included in U6.


Total persons on unemployment insurance fell by 215,000. I assume thats what he was referring to.


Total persons on unemployment insurance is not used to calculate unemployment rates by the government for the reasons you give. Two seperate statistics.


________


If you are looking for maybe a turning point....seasonally adjusted hourly wages and the SA average workweek both jumped in December. http://www.bls.gov/news.release/realer.nr0.htm

Net: The surveys driving U6 are crap.

The numbers aren't that tough. Doing this from memory but:


We are running about neutral on initial jobless claims vs attrition turnover, so we're not growing there. Net-neutral there on churn.

At best, we created 200,000 jobs last month in nonfarm payrolls (and that's dubious).

We had 215,000 who by definition fell off the unemployment rolls (therefore by definition are looking for work or underemployed as recently as 2-4 weeks ago), U6 should be about flat.

With 200k vs -215k... At BEST, U6 should be about neutral. That doesn't even factor in underemployed/deferred-employment retirees, yet.

U6 shouldn't be dropping. But it is.

Even the U6 survey system -- considered the more reliable of these -- is probably borked since it isn't catching that.

 

-----signature-----

I dont think you know if those 215k that fell off the insurance rolls did so because they found a job or b/c they ran out of benefits.


You are comparing different periods; +200k is December, -215k is one week in January.


You are using 3 different measurements, each with a margin of error and trying to jam them together. You are trying to do too much with limited data that measured slightly different things. Using longer time frames would also improve the accuracy of what you are tying to do.


Even great surveys will have a margin of error thats going to cause it to bounce around alot from month to month. Dont get too caught up in the month to month numbers.

Read again. The jobless claim and new non-farm payroll numbers I just cited SAY they didn't find a job.

 

-----signature-----