"Don't hurt people and don't take their stuff" - Matt Kibbe

5/11/10

Crunching The Numbers

I received a forwarded email this morning which contained an interesting analysis of the much talked about surge in employment the country has experienced according to the most recent government numbers. The analysis is by Arthur Cashin in his always informative and entertaining daily missive called "Cashin's Comments" which I enjoy immensely whenever I can get my grubby little hands on it.

His somewhat whimsical history lessons are mixed in with market commentary and analysis and I would happily pay him for a subscription to it. Unfortunately, it is not for sale and I am guessing it is available only to clients of his employer as a gift, and I am not among them. I receive it from time to time as a professional courtesy from a colleague in the financial industry and I don't use it as a tool to earn money, with so I'm guessing Art doesn't mind if I enjoy his efforts. At least I hope not.

I also hope I am not breaking any copyright laws by making it available. The copyright is held by UBS corporation and I do not know how to get their permission for a reprint. It is not an article which can be linked to as is my usual custom, so attribution is the best I can do. The following is his analysis:

Dissecting The Somewhat Suspect Payroll Data
"As we suggested in yesterday’s Comments, Friday’s non-farm payroll numbers got swallowed up by the on-going worries about Greece. But, the “jump” of 290,000 new jobs was a big topic on the weekend talk shows. There was a lot of “we’ve turned the corner” portrayals. Longtime readers know I’ve thought some of the improvement in the data was “suspect” (to be kind).

For the last eight weeks, Initial Unemployment Claims have averaged 450,000 per week. So, over the last four weeks, 1.8 million people were laid off. How does that fit in with the claim that 290,000 new jobs were created? The obvious answer is that it doesn’t.

So, let’s drill down into the payroll numbers to see what’s going on. The CES Birth/Death adjustment added 188,000 of those jobs. Birth/Death does not refer to people but to businesses. The BLS guesses how many new companies opened versus how many closed their doors. The BLS then uses that guess to guess again how many jobs those business created or lost.

Another 66,000 of the new jobs came from census hiring. Those are temporary jobs and those folks will be laid off later in the year. Speaking of temporary, another 26,000of the new jobs were non-census temporary. Let’s recap. A guess produced 188,000 of the jobs, 66,000 were census and 26,000 were temporary. Thus, it seems 280,000 of the 290,000 “new jobs” were either temporary or the result of guesswork. Some turn. Some corner."

It's always the "numbers behind the numbers" that matter if you are trying to get an accurate picture with statistics. It's why it took a few days to figure out that GM paid back all the money they owed to the federal government (themselves the owners of GM) ahead of schedule, with a different government loan. Which is also why a biased media and an intentionally misleading government don't want you to crunch them.

Instead of the numbers getting crunched, it may be the politicians who get crunched, at the polls.

1 comment:

Sara said...

The view from the ground still looks decidedly NOT improving on the job front. I have many friends who are underemployed and these aren't just people in the trades...I continue to be skeptical about the numbers I am hearing, although I appreciate this post for providing some backup to my uninformed skepticism.