By Ron Robins

This editorial is part of a series Mr. Robins* is writing on the ethics, and sometimes extraordinary biases he believes he has found behind the design and presentation of many government statistics. His view is that investors need to know much more about what is in these statistics in order to optimize their returns. His research reveals many surprises. For additional articles in this series, see Editorials. 

Revised and updated December 13, 2007

The business world waits with trepidation, the first Friday of each month, the release of the US unemployment/employment numbers. Stock, bond, currency and commodity markets often swing wildly with their release. The media focus on the numbers presented, and discuss their relevance to economic activity. But where is the analysis, the critique, of how these numbers are generated — or of their actual reliability?

Do all economists really believe that the US government’s unemployment data (and other statistics too) are beyond reproach? Are the big banks’ economists too afraid to dig into the numbers for fear of offending or confusing employers and clients? Where is the role of honesty, of ethical responsibility, to the publics these institutions serve?

Fortunately, discussion concerning the ethics and reliability of economic statistics does occasionally appear.

For instance, last year Philipp Bagus asserted in an article, The Problem of Accuracy of Economic Data, August 17, 2006 (http://www.mises.org/story/2280) “[That] we … face the question of why the problem of accuracy of economic data is rarely mentioned or passed over in silence in economics, while in the physical sciences this problem is widely acknowledged.” Further, “In contrast to physics, there is still no estimate of statistical error within economics. The various sources of error that come into play in the social sciences suggest that the error in economic observations is substantial… Economic statistics cannot be accepted at face value.”
In my research on US unemployment data, I have discovered some disquieting information. First of all, they concern the elimination of ‘discouraged workers,’ who used to be in the figures.

Discouraged workers are those who have been looking for employment for more than a year and have given-up looking for a job. They used to be included in the main unemployment numbers, but are now, conveniently left out! John Williams, statistician and economist, believes that when ‘discouraged’ workers and other ‘distorting factors’ are accounted for, then the true unemployment rate, measured in much the same way as it had been historically, would be closer to 12%! (See Welling@Weedon, February 21, 2006, Shadowing Reality interview with John Williams). At the time of Mr. Williams citing this, the US February 2006 unemployment rate was 4.7%, which is the same as for November 2007.

The second major concern is the inclusion in the non-seasonalized data — which influences the media headlined seasonally adjusted numbers — of escalating theoretically derived employment numbers from the business ‘Birth-Death Model.’ This model created by the US Bureau of Labor Statistics (BLS), tracks the net employment changes caused by business births and deaths.

Job gains in the Birth-Death Model have grown from less than half in 2004 to almost equalling the total employment gains in 2007? It begs the question as to how much of 2007’s employment gains are theoretically derived, and how much are real? The BLS appears silent on this point. With regard to the Birth-Death Model, the BLS states, “{The} BLS will continue researching alternative model-based techniques for the net birth/death component; it is likely to remain as the most problematic part of the estimation process.” Yes, it is certainly problematic.

The lack of analysis of jobs and other US economic data by mainstream economists and media is abysmal. Let economists and business journalists especially, take a lead in an illuminating debate around the make-up and ethics of such economic statistics. So far these individuals have really let down the publics they serve in this regard.

Note: This is his second article in an editorial series on Unethical Statistics. They concern the ethics surrounding the construction and portrayal of US government statistics.

© Ron Robins, 2006