Crunching the numbers to fit the PR box

Coincidentally with the need for a happy face on the president's economic performance, unemployment statistics got a boost from a mysterious statistical genie.

Since the latest series began in 1959, the participation rate in the labor pool -- the number of working age people looking for jobs -- has climbed steadily. Beginning in 2001, the number began to fall, and it has continued to fall. Absent this curiousity, 2005's unemployment rate would be 6.5 percent, rather than the fairly moderate 5.1.

Something smells when the current administration can only claim the addition of 4.8 million jobs over five years, while adding 1.9 million to the rolls of the unemployed and still get something to brag about on the nightly news. Clinton's eight years saw 18.4 million new jobs, and a subtraction of 3.9 unemployed.

The labor pool has grown inexorably since statistics have been kept. It tends to react a bit with the availability of jobs, it's a fact, but not this much. One of the wonders of employment statistics is that even as the denominator has grown, under Democratic presidents the unemployment rate has consistently dropped from one year to the next.

The Bush II collapse of the labor pool is, however, unprecedented. While the participation rate has come down a tenth of a point here and there, it has come down more than one-tenth only four times since 1971 -- in 2001, 2002, 2003 and 2004.

I began to look closer at this when I caught on to John Williams pieces on how official numbers are crunched until they fit the PR boxes. I admit I don't know exactly the methodological gymnastics that produced these particular participation rate anomolies, but they are there, and they're in the breakouts for the state, too. Maybe I'll subscribe to Williams' newsletter for us.