Fun Facts

A few weeks ago, I wrote a piece about U.S. Bureau of Labor Statistics jobs reports whose executive summary headlines were not substantiated by underlying data in the studies’ tables.     

Since then, many others have begun to notice the apparent disconnect between headlines and data.  So it is no surprise that a virtual firestorm has erupted over the latest jobs report released by BLS (Employment Situation Summary), which opens with this terrific announcement:  

“Total nonfarm payroll employment rose by 243,000 in January, and the unemployment rate decreased to 8.3 percent, the U.S. Bureau of Labor Statistics reported today.”

President Obama quit reading right there and called a press conference.  Mitt Romney quit reading right there and spoke to the media.  Santorum and Gingrich quit reading right there and said Mr. Romney no longer was needed to fix the economy.  The media quit reading right there and ran with the story that Mr. Obama will now surely be re-elected.  Everyone, it seems, quit reading right there. 

Well, not everyone - Ron Paul, I suspect, never read it in the first place, nor did I.  But if you are going to read it, don’t stop right there - because it gets very interesting very quickly if you keep going.  Like a pack of cigarettes, this month’s employment report comes with its own Surgeon General’s warning: 

“Data users are cautioned that these annual population adjustments affect the comparability of household data series over time.”  

Not to be a jerk, but when BLS says employment “rose” in January, they mean rose from December, which is...um…comparing the data series over time.  Just sayin’.  And what adjustments are they talking about?  

“The (benchmark) adjustment increased the estimated size of the civilian non-institutional population in December by 1,510,000, the civilian labor force by 258,000, employment by 216,000, unemployment by 42,000, and persons not in the labor force by 1,252,000.”

Catch that? 216,000 of those 243,000 jobs “created” in January were the result of “benchmark adjustments” - BLS channeling their inner Ben Bernanke and quantitatively easing 1.5 million new fiat people into existence while putting the digital whiteout to 1.2 million other Americans.  Too bad statistical methodologies are not the economy; it would be a lot easier to fix. 

Most Americans don’t care about methodology; we just want to know whether or not actual unemployment was reduced in January.  Here is how BLS itself described the situation for the various categories of the unemployed:

“The number of long-term unemployed was little changed at 5.5 million…the number of persons employed part time for economic reasons, at 8.2 million, changed little in January…2.8 million persons were marginally attached to the labor force, essentially unchanged…there were 1.1 million discouraged workers in January, little different than a year ago…”

Little-changed, changed-little, essentially-unchanged, and little-different don’t sound like change you can believe in if you are one of those millions hoping for things to change.  The one category that did go down in January was the number of unemployed who lost their jobs since Christmas and could not find new ones – reported at “only” 2.5 million. 

And why does that particular segment improve so suddenly?  Because black unemployment fell by an astounding 14% in January!  Seriously - I told you it would get interesting.  At first I thought NBA, but the lockout ended in December, and it’s really not that many guys.  Hispanics did well too, according to BLS, with a reduction in their unemployment rate that is five times greater than that the 1% improvement for whites. 

So if we are to believe BLS data at face value, then the economy boomed in January because the private sector went on a one-month minority hiring binge unprecedented in all of human history.  Inexplicably, MSNBC did not interrupt their interruptions to inform us that racism has ended. 

What do you think about that, Milwaukee – are the good times rolling on North Avenue?  Are you buying what BLS is selling?  It sure doesn’t feel like black unemployment fell sharply in January, and that is definitely not what they were talking about on the local Sunday talk shows yesterday.       

Which brings us back to why BLS specifically warned us not to take those 243,000 new jobs to the bank – the economy didn’t actually recover.  They simply adjusted their benchmarks, whatever the heck that means, and the resulting statistical windfall, for reasons none of us could possibly understand, fell disproportionately to the computations of unemployment rates for minorities.    

But that still leaves a gain of 27,000 non-benchmark-adjusted new jobs; that’s a good thing, right?  The BLS tells what to make of the number in their methodology disclosure:    

“…the 90-percent confidence interval for the monthly change in unemployment as measured by the household survey is about +/- 280,000…the threshold for a statistically significant change in the household survey is about 400,000.”

In layman’s terms, disregard. 

Politicians and their supporters should take extreme caution when using BLS data to take credit for job creation or assign blame for job loss.  Using statistical forecasting methods to estimate past employment is a strange way to count jobs anyway.  For the life of me I don’t know why the government relies on phone surveys instead of compiling employment data from IRS records. Every time someone is hired or terminated the IRS is notified; we don’t need to guess about it. 

But the silver lining in all this is that I learned something new.  So after benchmark-adjusting myself, I can report with a 90% confidence interval of +/- 120, that I lost 60 pounds in January.  Sure beats working out…


“Moment Of Clarity” is a weekly commentary by Libertarian writer and speaker Tim Nerenz, Ph.D.  Visit Tim’s website www.timnerenz.com to find your moment and order Tim’s new book, “BRING IT!”