Lost Bet on Unemployment

Mike Shedlock
|
Posted: Oct 08, 2012 12:01 AM

On Friday, I lost a bet made in March of 2010 regarding unemployment. I bet that the unemployment rate would not dip below 8% before June 2015.

I crunched numbers many ways and simply decided there is no way the economy could possibly grow enough jobs. It didn't and still won't.

I posted this chart at the time.
 
At the height of the housing bubble, the economy only added 212,000 jobs a month. I figured we would not come close to that, yet even if we did, that still would not be enough.

A miscalculation got into my way, otherwise known as a plunge in the participation rate.  I knew full well the participation rate would drop on account of boomer demographics. But I never expected the plunge we got.

Monthly Job Growth 1999-2009



Were it not for millions dropping out of the labor force, the unemployment rate, even with the September barrage in part-time jobs would still be over 10%.

Had I made a similar bet on employment, rather than unemployment, the bet would still be going on. Interestingly, that bet would not have seemed as good to me at the time because the economy is naturally growing and employment with it.

Total Nonfarm Employment



Employment Just About at June 2005 Level

As you can see, employment has only recovered to a level seen in 2005. That is in spite of the fact the worker population expands every month (at least in theory). Bernanke thinks it takes 100,000-125,000 jobs a month to keep the unemployment rate steady.

I think because of demographics, it probably only takes 75,000 jobs a month.

However, at some point in a recovery, the participation rate starts to rise as discouraged workers return to the work force. Accounting for the return of those discouraged workers to the job force, assume it takes 100,000 jobs a month to hold the unemployment rate steady.

That amounts to 1.2 million per year and 3.6 million in three years (3.9 million since the recession ended).

Month-by-Month Employment Totals (in Thousands)

Month Employment
2010-01-01 129279
2010-02-01 129244
2010-03-01 129433
2010-04-01 129672
2010-05-01 130188
2010-06-01 130021
2010-07-01 129963
2010-08-01 129912
2010-09-01 129885
2010-10-01 130105
2010-11-01 130226
2010-12-01 130346
2011-01-01 130456
2011-02-01 130676
2011-03-01 130922
2011-04-01 131173
2011-05-01 131227
2011-06-01 131311
2011-07-01 131407
2011-08-01 131492
2011-09-01 131694
2011-10-01 131806
2011-11-01 131963
2011-12-01 132186
2012-01-01 132461
2012-02-01 132720
2012-03-01 132863
2012-04-01 132931
2012-05-01 133018
2012-06-01 133063
2012-07-01 133244
2012-08-01 133386
2012-09-01 133500


Number Flashback

  • Jun 2009 employment: 130,503,000
  • Jan 2010 employment: 129,279,000
  • Sep 2012 employment: 133,500,000
  • Jun 2005 employment: 133,607,000

Numbers and chart from Fred - St. Louis Fed - PAYEMS.

The recession ended in June of 2009, and here we are, over three years into a recovery, with actual employment below where it was in June of 2005.

Starting from the end of the recession, and looking to hold the unemployment rate flat since then (assuming 100,000 jobs per month required), employment would need to be at 130.503 million + 3.9 million (a total of 134,403 million). As you can see, we are about 900,000 jobs short!

Statistically speaking, the expectation is the unemployment rate should have risen from the level it was at in June 2009. It would have, except for the amazing drop in the participation rate. In terms of employment, I had the right idea but made the wrong bet. Bryan Caplan wins.

Mike "Mish" Shedlock
http://globaleconomicanalysis.blogspot.com