re: It's that time of the month again - unemployment/jobs reports in the morning

quote:

---

what am I missing

---

Nothing. Since day 1, this is how the administration reports job numbers. They count on the public and reporters to not understand basic mathematics.

quote:

---

Job numbers increased slightly, yet again, and not enough to keep up with the population rate entering the workforce.

---

Take a look at a very revealing pair of numbers from today's report that no one tends to notice:

"After accounting for the annual adjustment to the population controls, the civilian labor
force rose by 499,000 in January"

"Total employment, as measured by the household survey, increased by 616,000 over
the month"

The various annual adjustments made by the BLS make for some interesting results.

Today's report (Table A) shows a population increase of 170k from December to January. It also shows an increase in the labor force of 523k and an increase in the number of employed of 638k for the same December-to-January period.

So only 170k people were added to the population month-over-month, yet the labor force increased by 523k and the number of employed persons increased by 638k. Awesome !

Oh, wait.

We now have to take into account the new January 2014 "population control effects".

What effect did they have on the numbers in the table?

Lowered the population increase by 2k, to 168k.
Lowered the labor force increase by 24k, to 499k.
Lowered the number of employed persons by 22k, to 616k.
As you can see, these adjustments bring the Table A numbers back in line with the report's narrative numbers.

Allowing for the population increase, that would indicate that 331k folks re-entered the workforce in January (499-168).

To reach an increase of 616k in the number of employed persons, the entire number of persons who re-entered the workforce would have had to find jobs in January, as well as 285k additional persons who were not employed in December. And in addition to that, the equation would mean that anyone who lost a job between the Dec. and Jan,. reporting periods would have already found a replacement job, which we know is a ridiculous assumption.

Yet the BLS report shows an increase of 113k jobs for the month.

This illustrates the foolishness of releasing the "unemployment rate report" and the "jobs report" together each month.

There is no direct correlation between the two, which is why the numbers so often result in complete confusion.

The Household Survey numbers are junk. Excellent example of "garbage in, garbage out". Poll one out of every 2,300+/- households and use very loose parameters (like working one hour the week before makes one "employed"), and allow for the respondents truthfulness/accuracy in answering the questions, as well CB employee "prompting", and one arrives at a rate that is measurable to the tenth of a percent. Throw in the "adjustments" (seasonal, annual, population, etc.) and the numbers become ballpark figures at best. Yet every change of a tenth of a percent in either direction becomes news.

If I'm Obama, I'm in Michigan today saying, "Hey folks, did you see where 616,000 more Americans are are employed today than last month?", and wait for the cheers from a crowd that has no clue where that number came from and how meaningless it really is.

Indeed. The birth-death model has been dysfunctional for some time.

The seasonal adjustment can wreck havoc during the holiday season if hiring isn't in line with previous years as well...

Of course there is little appetite to explore, nor curiosity from our betters in the media. Soundbites shall suffice.