Washington Times chief political correspondent Donald Lambro based his recent assessment of the American job market on a completely different measure than the one he used ten years ago.
In 1994, Lambro touted the results of the Bureau of Labor Statistics' (BLS) payroll survey as "far more reliable" than the BLS's household survey. The payroll survey collects information from the payroll records of approximately "160,000 businesses and government agencies covering approximately 400,000 individual worksites," according to the BLS, and BLS commissioner Kathleen P. Utgoff has called it "the best indicator of current job trends." In 1994, the payroll data was less favorable than the household data for President Bill Clinton; but in his October 21, 2004, column, Lambro denigrated the reliability of the payroll survey, this time heralding the results of the BLS household survey -- which now offers a more favorable employment outlook for the Bush administration than the payroll survey does.
As New York Times columnist Paul Krugman noted on August 10, Federal Reserve Board chairman Alan Greenspan said (registration required) of the payroll survey, "Everything we've looked at suggests that it's the payroll data which are the series which you have to follow." As the Chicago Tribune reported on August 7, the household survey queries only about 60,000 households and is consequently "considered less reliable."
From Lambro's October 21, 2004, column, titled "Penetrating the economic fog":
As for overall employment, it's not true, as Mr. Kerry states, that Mr. Bush has presided over an economy with a net loss of jobs over the past four years. The Labor Department's household survey, providing a more accurate gauge of how many people are working, shows total employment has grown by 3 million. The payroll report does not pick up most of these people because many are self-employed and work as independent contractors -- a major and continuing trend in our ever-restructuring economy.
From Lambro's June 11, 1994, column, titled "Beginning to feel the curse of Clintonomics":
The government's household survey showed 250,000 jobs were created last month. But the far more reliable, though less noticed, payroll survey showed only 191,000 jobs were added to payrolls, and 70,000 of them were due to the return of striking truck drivers.