Economic snapshot | Wages Incomes and Wealth

The ‘sad story’ of the current employment picture: 4.2 million jobs below normal

Share this page:

See Snapshots Archive.

Snapshot for October 25, 2004.

The ‘sad story’ of the current employment picture


4.2 million jobs below normal

The unprecedented loss of payroll jobs 42 months after the start of the last recession is well known.  However, it has been argued that the two surveys of employment conditions (the payroll survey and the household survey) tell conflicting stories.1  Some have claimed that the household survey paints a much more favorable employment picture.  Last spring, however, two economists at the Cleveland Federal Reserve issued a study entitled “Employment Surveys Are Telling the Same (Sad) Story.”2   This Snapshot adds six more months of data to their central figure and finds that the household survey continues to tell “the same sad story” as the payroll survey.

The economy should be adding about 150,000 payroll jobs a month just to keep up with growth in the working-age population. The labor market develops more slack every month that fewer than 150,000 jobs are created.  In a normal recovery, the economy would be adding 300,000 payroll jobs a month (see JobWatch.org for further discussion of these calculations).

When employment grows as fast as the population, we have a stable “employment-to-population ratio,” or “employment rate.”  The employment rate typically falls not only during the recession, but also for several months afterward.  In the previous seven recession/recovery cycles (that lasted at least 42 months), the employment rate typically fell for a year and a half after the recession began.  Then jobs begin to grow much faster than the population and the employment rate rose.  In the average of prior cycles, the employment rate had fully recovered by the 44th month after the start of the recession.

As the figure below shows, the trend for the first 18 months of the current cycle closely resembles the average of previous cycles.  This cycle did not diverge from the past pattern until after September 2002—well after the stock market had tumbled and both September 11 and corporate accounting scandals had come and gone.  As the figure reveals, the employment rate continued to fall through the spring of 2003 and has effectively remained flat since then.  Yes, employment has grown in the last two years, but not enough to tighten the labor market. 

Employment-to-population ratio, change from business cycle peak

When the recession began in March 2001, the employment rate stood at 64.3%.  By last month, it had fallen by two percentage points to 62.3%.  In contrast, a normal jobs recovery would have brought the employment rate back to within 0.1% of the initial employment rate, or up to 64.2%.  Today there is a working-age population of 223.9 million.  Thus, if we had experienced a normal recovery, we would have a labor market almost as tight as when the recession began, and 4.2 million more people would be employed. 

This Snapshot was written by EPI Research Director Lee Price with research assistance from Yulia Fungard.

1. Both John Snow on October 10 and Don Evans on October 17 touted the household survey numbers on “CNN Late Edition with Wolf Blitzer.”  Elaine Chao did the same August 6 on CNNFN.  Also see Tim Kane, “The Myth of a Jobless Recovery,” Heritage Foundation, March 25, 2004.  For an analysis of the misuse of the household survey, see Elise Gould, Measuring Employment Since the Recovery: A Comparison of the Household and Payroll Surveys, EPI Briefing Paper, December 12, 2003.

2. Mark Schweitzer and Guhan Venkatu, Research Department, Federal Reserve Bank of Cleveland, May 15, 2004.


See related work on Wages Incomes and Wealth

See more work by Lee Price