Back to top Jump to featured resources
Article filed under The Public Good

American Voice 2004: Who’s right and who’s wrong in the job growth debate?

| Written by David Morris | No Comments | Updated on Jun 1, 2004 The content that follows was originally published on the Institute for Local Self-Reliance website at https://ilsr.org/whos-right-and-whos-wrong-in-the-job-growth-debate/

Q.   Republicans insist that during George W. Bush’s first three years in office several hundred thousand jobs have been created. Democrats maintain that during this same time period several million jobs have been lost! Both are relying on government data. How can that be? Who is right?

Answer:

They’re both right, although one may be righter than the other. The Department of Labor’s Bureau of Labor Statistics, the primary source for employment information, uses two very different methods to calculate job growth.

The payroll survey relies on monthly unemployment insurance records kept by 400,000 companies. The household survey relies on information from monthly interviews of 60,000 households.

The two methods can produce very different results. The February payroll survey, for example, estimates that 2 million jobs have been lost since 2001. The household survey, conducted the same month, concludes that the economy has gained more than 500,000 jobs.

Which is more right? According to recent Congressional testimony by Alan Greenspan, whose reappointment as Chairman of the Federal Reserve President Bush supports, “We have concluded that the data on the so-called payrolls survey is surely the more accurate of the two…” The payroll survey samples 600 times as many workers as the household survey.

The household survey is useful because it counts self-employed individuals, an important and growing part of the labor force that is not counted in the payroll survey. Greenspan offers this bit of sound advice, “So if you add both of them together you’ve probably got the best estimate of what true employment is, which”, he points out, “still shows something far more closely related to the payroll rather than the household survey.”