Economic Old Age: 40% Unemployment Ain’t Awesome

Last time it was the 100-year "contagion" flood that supposedly upended the Fed's "Goldilocks economy". As shown in the second chart above, the latter was smoothly chugging into the Keynesian nirvana of U-3 full-employment at the end of 2007, and on a pathway virtually identical to the present.

It is more than evident, of course, that Wall Street desperately wishes to be fooled twice by the Goldilocks postulate, and that's why the trading bots will thrash around inside their chart points and moving averages until this cycle's version of the Black, Orange or Red Swan , as the case may be, makes its unscheduled appearance.

In the interim, however, the Awesome Economy narrative gets more threadbare by the month. As Jeff Snider astutely observed in his commentary on today's report---at some point it gets pretty hard to hide 16.6 million missing workers.

What he means is that if the labor force participation rate had not plunged from more than 67.0% to the 62.8% level reported again for April, there would be 16.6 million more persons employed in the US economy today at the ostensible 3.9% unemployment rate.

The Keynesian gummers reject Snider's point entirely, of course, on the vague theory that retirements and the aging demographics of the US explain away much of the change in the participation rate.

As a matter of fact, they don't. And, besides, the whole BLS employment/unemployment reporting framework and model is essentially a pile of garbage that might have been relevant during the days of your grandfather's economy, if even then.

That is, it is built on the flawed notion that labor inputs can be accurately measured by a unit called a "job" and that an economic trend in motion tends to stay in motion.

To the contrary, in today's world labor is procured by the hour and by the gig---meaning that the "job" units counted in both the establishment and household surveys are a case of apples, oranges and kumquats. The household survey, for example, would count as equally "employed" a person holding:

  •  a 10-hour per week gig with no benefits;
  •  a worker holding three part-time jobs adding to less than 36 hours per week with some benefits; and
  •  a 50-hour per week manufacturing job (with overtime) providing a Cadillac style benefit package.

Beyond that, the underlying monthly surveys are tiny, primitive and utterly lacking in quality control among respondents. As a result, the statistical wizards at the BLS smooth it all over with ad hoc adjustments and guesstimates (e.g. the notorious birth-death model) and trend-cycle statistical models that essentially project into the current month the statistical trend line then underway.

In short, the monthly jobs report is not an accurate empirical snapshot of where the real world labor market actually is; it's a modeled projection of where it should be.

So when Hampton Pearson (CNBC) reports the number as coming in where it should be, the bobble-heads all wag enthusiastically about another "solid" jobs report.

And that's also why the BLS jobs confection is useless at turning points in the business cycle. During the 2008-2009 employment collapse, for instance, it over-reported the monthly level by nearly 500,000 jobs per month.

View single page >> |
How did you like this article? Let us know so we can better customize your reading experience.


Leave a comment to automatically be entered into our contest to win a free Echo Show.
Gary Anderson 3 years ago Contributor's comment

Without increases in welfare things would have been worse. I think the Republican guerrilla in you wants to cut social programs to speed up a revolution.