Thanks to


Last week, past GE-CEO Jack Welch created a bit of a stir when he tweeted that the just-released unemployment figures were, “Unbelievable jobs numbers..these Chicago guys will do anything..can’t debate so change numbers,”

Many people were amazed that with only 114,000 jobs added to the economy, the unemployment rate could drop 0.3% to 7.8%.  The average job growth each month in 2012 has been 146,000, yet the unemployment rate has remained between 8.1 and 8.3%.   It seems strange that less than average job growth computes to far greater than average lowering of the unemployment rate – in fact, the second biggest one month drop in unemployment in at least the past 10 years.

The timing, just before the election, made many people question.  I was one.  But, I found it difficult to believe that a President, or his minions, could get the cooperation of hundreds of bureaucrats to ‘fudge’ the numbers.  Most folks agreed with that view and made light of the ‘conspiracy theory’.


Thanks to

I went to the BLS website to learn what I could.  It was very enlightening.  First, I learned that the “Household Unemployment Rate” is the reported unemployment rate that has been in the news so much for the past ten years.  It is determined exclusively from phone surveys completed by employees of the Census Bureau.  Some 60,000 surveys (calls) are made by 2,200 BLS Surveyors during the month.  I wonder, since this is all verbal, could a surveyor’s politics ‘interpret’ answers to fit his or her views?  Could this sway the numbers?

Next I saw an announcement that the Employment Situation report for October 2012, will be released on November 2, 2012, the Friday before the Election.  At that time they would also do any revision to the numbers of the previous month.  That report could be more important than the September report was since it is just 4 days before most people vote.  If there is a conspiracy rather than a cultural bias among surveying employees of the BLS, wouldn’t they wait until November 2 to have the most effect?

Then I found that between September of 2011 and September of 2012, the number of people reported as “Not in the Labor Force” increased by 2,643,000.  (“Not in Labor Force” went from 86,067,000 to 88, 710,000.)  I highly recommend that you go to the BLS site and look at Table A which is the summary of the statistics.  See if you can understand why most of the statistics make it look like unemployment seems worse but, the one-most-reported statistic looks better.  I can’t.

I also ran into an article in Media Matters about the Welch Tweet that was interesting.  Of interest was the comment by one Michael Doan who said,

“-If you were fired Monday and got the census call later that week, you are counted as employed (part-time) since you worked, just not 35 hours.
-If you gave up over 1 month ago for any reason whatsoever, including perceived discrimination, you are classified as Not Part of the Labor Force – meaning statistically, you’re a 3-year-old girl that can’t be employed and can’t be included in the statistic.
-Looking through online/newspaper classified ads doesn’t count as actively seeking employment – but looking at online union registers does.  
-If you are seeking work for any time beyond the current day (a week before graduating, for example) and can’t find work for that day, you are not considered unemployed.
-Being homeless and working 15 hours per week washing windows at stoplights means you are employed.
-Those surveyed are never asked if they are unemployed directly”

I find it very hard to believe that these statistics can be swayed by political force, however (see my post, After the “But”) I don’t think it would be impossible.  More likely, the numbers are swayed by the political leanings of the surveyors, much like major media news reporting is swayed by the politics of the reporters.  Any political lean of the unemployment numbers would more likely be as a result of individuals (government employed survey takers) rather than a broad conspiracy of Mr. Obama or his reelection team..

I also think that the statistics are so confusing as to be almost worthless in the short term.  Maybe over a long historical perspective, there is value, but for now I think they contribute most to political posturing.