NCPA - National Center for Policy Analysis

Employment Data Questioned

July 10, 2001


Each month the Labor Department releases much-anticipated figures on the unemployment rate and change in the number of jobs on company payrolls. The figures were 4.5 percent and minus 114,000 jobs, respectively, for June.

But some economists charge the figures are more than suspect: they say the figures are false, and understate the slowdown in employment.

Here's why:

  • Although the Labor Department reported payroll employment fell 114,000 in June, that includes a "bias adjustment factor" that adds about 160,000 jobs a month -- which is supposed to capture employment in newly started firms the department misses in its survey.
  • The department really didn't know how many new jobs were created by start-up companies, so it assumed 155,000 for June.
  • The problem is that when the economy slumps, so do new business start-ups -- and a good indicator of these is the Conference Board's index of help-wanted advertising.
  • This index has plummeted back to levels last seen at the end of the 1990 recession -- at which time the bias factor fell to zero.

Critics also question the department's household survey -- which asks people, rather than businesses, about layoffs.

  • Over the past five months, the household survey shows a fall in employment of over 1 million.
  • But over the same period, the published payroll survey has fallen only 45,000.
  • However, if one removes the monthly addition of 155,000 from the bias factor, payroll employment would be down 269,000 in June and 872,000 over the past five months.

Making these adjustments, it would appear that the nation has experienced a bigger drop in employment than occurred over the first five months of the 1990 recession.

Source: Lincoln Anderson (LPL Financial Services), "Labor Statistics Are Lying," Wall Street Journal, July 10, 2001.

For text (interactive subscription required)


Browse more articles on Economic Issues