Employment and Unemployment numbers may be dead wrong

I have some difficulty with Government Data (which, indeed, may be an oxymoron). So, when the Employment/Unemployment news release came out from the Department of Labor, Bureau of LAbor Statistics (BLS) last week, two things bothered me.

First, that the Total Non-AG Employment dropped “only” -247,000. The news media picked up on this, as did the talking head economists on TV, even going so far as to say that the new numbers clearly show that “The recession is officially over.” Now, before I go on, let me assure you that I believe that any GOOD news can have a positive impact on the consumer, which could be good for the economy. However, GOOD NEWS, that can be misleading, carries with it the responsibility to point out risks associated with the news. In this case, the Bureau of Labor Statistics (Department of Labor) failed to tell the public of the problem they have with their own data in the news release.

So, I called the BLS and got information on their methodology and the margin of error. Sure enough, when you read the small print, the margin of error for the Non-Ag Total Employment number is +- 430,000.

So the BLS is 90% certain that job growth was -677,000 or + 183,000 or somewhere in between, and, of course, since the margin of error includes ZERO, the news release states that they do not know whether there was an increase or decrease. Not withstanding their conclusion that they cannot determine whether the number went up or down, the governmental agency issued a news release that stated that there was a job loss, and the loss was specifically, 247,000 jobs, which is about ½ of the previous loss rate. It is mind-boggling to me that an agency of the federal government would make up a number and release that number as fact. Now is a time in our economic history when these numbers are more critical than ever, and making investment decisions, based on these numbers is extremely risky. Why would they do that?

The second thing that bothered me was the fact that, even though the annualized, estimated job loss rate (-247,000 X 12) is about -3,000,000 jobs per year (if the data were accurate), the BLS release said the Unemployment Rate went DOWN from 9.5% to 9.4%. The talking heads jumped all over that number and the stock market went UP. That just does not make any sense. In the summer, the Unemployment Rate usually goes UP, even in good times. When we are losing 3 million jobs a year, how does the Unemployment Rate go DOWN?

The same accuracy issue applies to the Unemployment Rate news release. The reported change was -0.1%, as they announced that the Unemployment Rate dropped from 9.5% to 9.4%. The margin of error, however, is +- 0.19%, so the 90% certainty range, from +0.9% to -2.9%, once again, includes ZERO. So the reported change of -.1% is statistically insignificant, and they do not know whether there was an increase or decrease. Even so, they issue a news release that states specifically that they know that the Unemployment Rate went DOWN.

So, the news release comes out, the talking heads talk and move on, the RECESSION IS OVER, and everything is just FINE, and no one says, “Wait a minute, that’s not right!.” No one issues a warning that the numbers could be misleading and that making investment or business decisions based on these numbers could lead to financial disaster. Are we so numb and so brain dead that we just sit there and let the Federal Government tell us what they want and we don’t care?

I have included a link to the BLS News Release, below. Once on the BLS page, click the .pdf version of the news release.

http://www.bls.gov/news.release/empsit.nr0.htm

I direct your attention to page one (1) for the news and page eight (8) under the heading "Reliability of the estimates" for the discussion of the error factor.