Jobs created or saved

Posted by Marc Hodak on February 26, 2009 under Irrationality | Be the First to Comment

For me, the most surreal aspect of the press’s credulous reporting on the unconstrained orgy of spending they insisted on calling the “stimulus package” was their reporting of “jobs impact.”  The scare quotes and accompanying skepticism arise from the definition of “jobs impact” as “jobs created or saved,” an unverifiable notion specifically designed to eliminate any accountability for the numbers being offered.

Nevertheless, those numbers were dutifully published.  In the Houston Business Journal,  The Tennesean, New Mexico news station KRQE, etc., everyone reported the local number of “jobs created or saved” by the spending bill.  CBS regurgitated these spoon-fed numbers:

California…will see a “jobs impact” of 396,000.  Texas (with 269,000 jobs predicted), New York (with 215,000 jobs predicted) and Florida (with 206,000 jobs predicted) also fare well.

Less populous states see far less impact: Alaska, North Dakota, Vermont and Wyoming are each only predicted to see a “jobs impact” of 8,000.

Hmm, 8,000 each for the smaller states.  What a coincidence.

I wonder if the CBS reporter noticed that each of those states had exactly one congressional district?  Or that when you looked at each of the districts and states, as reported in the various local journals, “jobs impact” was consistently close to 8,000.  So, if you liked arithmetic and wanted to know the number of jobs created by state, you could simply multiply the number of congressional districts in the state by 8,000.

So, let’s just flick on our brains here for a moment, and ask ourselves:  What are the odds are that any given district will see a “jobs impact” of about 8,000?  Statistically speaking, the standard deviation around the mean of 8,000 is probably in excess of, oh, 8,000.  In other words, the district-by-district reporting is totally bogus.  Which makes the state-by-state estimates pretty bogus.

The inventors of the “jobs impact” numbers were looking at the total estimated jobs created from the whole bill and dividing by the number of districts to get their ludicrous district- and state-level results.

How bogus is the aggregate national number?  Mark Zandi, economist to the star politicians, noted with typical humility:

Yes, there’s a high level of uncertainty. But my estimates are as good as you’re going to get, and they’re good enough to be useful in trying to evaluate whether we should do this or not.

So, the answer is:  bogus, all the way up, as well as down.  It’s as if they passed a random bill for $535 million, and advertised that they were spending $1 million in every district across the land.

So, why did the press go for Obama’s uncannily even district and state “jobs” estimates?  Why did they participate in conspiring to create an aura of precision that so obviously wasn’t justified?

I don’t buy into conspiracy theories, so I won’t simply assert that to ask the question is to answer it.  I almost prefer to say that most of the press is as innumerate as they are literate.  Almost.