Friday, July 10, 2009

The "Exhaustion Rate" Underestimates the Issue

There is some confusion out there as to what exactly the "exhaustion rate" is (I had been confused myself). The following definition has been commonly used by many blogs and media pundits alike:

The United States Department of Labor publishes statistics on the "exhaustion rate" - this is a measure of the number of people who have used up their benefits, and will no longer be receiving unemployment checks.
This definition had been correct. And if this were still true, the chart below would show just how large a crisis we were in with those that are no longer able to receive unemployment at almost 50% of their "unemployment" class.

Unfortunately, this equation misses a large change in the U.S. unemployment legislature which results in the figure actually being even worse.

Again, the definition HAD been true. The reason why it is HAD, rather than HAS is that the calculation for the exhaustion rate has not changed, even though benefits have been extended not once, but twice over the past year (first from 6 months to 9 months in July and then to 12 months in November).

Why is this important? Because the Department of Labor never changed their calculation. The calculation is as follows:
  • The 12 Month Average of those Receiving their Final Payment divided by...
  • The 12 Month Average of those Receiving their 1st Payment, with a 6 Month Lag
The 6 month lag was put in to capture the number of those in the "unemployment class" that were now receiving their final payment. Lets go to an example assuming unemployment benefits could only be collected for 6 months:
  • 100 people became unemployed 6 months ago
  • 50 people six months later were receiving their final payment
  • The exhaustion rate equals... 50%
Using that calculation, the number of each is as follow...

But, the length was extended. So, rather than 100 people becoming unemployed 6 months ago, that number is more in the neighborhood of 85 becoming unemployed 12 months ago (the number of those becoming unemployed has grown) AND unfortunately the 50 people receiving their final payment is correct, but it was after 12 months of collections.

And that is exactly what has happened. Swapping in a 9 month delay after July and a 1 year delay after November, we get the following.

So that ugly 49% exhaustion rate figure? It's more likely at 56% and I can only guess where June will come in at.

Source: Department of Labor


  1. Was it Calculated Risk who put up a chart showing non-farm employment as percent of population since 1970 or so?

    David Rosenberg has been translating the contraction in hours worked into full-time-worker equivalents and adding that to unemployment numbers, coming up with total job destruction rates of around 700-800,000 per month.

  2. Interesting... please post David's chart if you can find it.

    Here is something interesting and along the line of no growth in jobs (but not as a % of the population).

  3. Also, forgot about this...

    'The End of the Private Sector Boom'...

  4. Jake, you can sign up for Rosenberg's Daily research notes at Gluskin Sheff's website.