August 12, 2009
Last Friday, a Bloomberg.com headline read: “U.S. Stocks Gain, Treasuries Drop as Unemployment Rate Declines.”
Let’s have a look at the reported decline in the rate of unemployment. Do you believe that the U.S. auto industry added 28,000 jobs in July amidst the GM bankruptcy, sell-off and close-down of GM auto divisions, and demise of GM suppliers? No? Well, that’s what the Bureau of Labor Statistics reported.
The 28,000 new jobs were created by “seasonal adjustments.” July is a month when jobs are automatically added by the BLS to seasonally smooth the layoffs of autoworkers during July’s retooling for the new model year. This year, most of the retooling did not occur, yet the annual seasonal adjustments did. Adjustments are also made for supporting industries, which are partially idled while auto production halts for retooling.
More phantom jobs were created by the “Birth-Death Model.” The payroll jobs data contains guesses about the numbers of new startup company hires and jobs lost from business failures. Failed businesses don’t report the lost jobs (deaths), and new jobs from startups (births) are not captured in the reporting. The government estimates these numbers, but the estimates are based mainly on growth periods, not on recessionary times. Consequently, during economic downturns, the Birth-Death Model overestimates the number of new startup jobs and underestimates the job loss.
The employment outlook was further improved by pushing another cadre of workers, who have been unemployed for too long, off the unemployment rolls. Remember that since the Clinton administration, the long-term discouraged (people out of work for more than one year) are not counted as being in the workforce. The length of the current downturn means that short-term discouraged workers, who are counted among the unemployed, are now moving into the long-term discouraged category, which simply erases their existence and lowers the measured rate of unemployment.
All sorts of distortions can find their way into the official statistics. For example, industrial production estimates are based on electricity consumption. Unusually hot weather, which causes a jump in air conditioning use, appears in the statistics as an increase in industrial output. Cool weather spells during summer reduces electricity use and results in a phantom drop in industrial output.
Nominal retail sales figures can increase from an uptick in inflation.
An increase in real gross domestic product can be the result of underestimating inflation.
Other distortions come from the year-to-year comparisons. As time passes, new comparisons are no longer with previous peaks, but with more recent lows. Thus, reported declines are less severe than previously, which makes things sound better when they aren’t.
By spinning the financial news, the appearance of recovery is created, and this lures people back into the stock and real-estate markets where they can lose the remainder of their wealth.