Growth in full-time employment lags previous business cycles
By Elise Gould with assistance from Gabriela Prudencio
According to today's report from the Bureau of Labor Statistics, total full-time employment was 117.2 million and part-time employment was 24.5 million in June 2005 (see data note below). The vast majority of U.S. employment (83%) is in full-time work, but unfortunately growth in full-time work in this business cycles is weaker than that of previous ones.
Figure A shows full- and part-time employment since the start of the most recent recession in March 2001. Full-time job creation and destruction is more stable over time than part-time employment. Following the economic peak in March 2001, full-time employment in the household survey fell until June 2002, and it took 40 months (July 2004) to regain its pre-recession level.1 Since the peak, full-time employment has grown by 2.4% or 2.6 million.
The trend in part-time work has been substantially different. After an initial decline, part-time employment increased by 5.4%, or 1.2 million, but after about two years it leveled out.
But how do these trends compare to other business cycles? Figures B and C compare the 51-month period following the peak (i.e., March 2001 to June 2005) to the same length periods following the 1990 and 1981 recessions.
In the recession which began in July 1981, the U.S. economy experienced a larger decline in full-time employment accompanied by a quicker recovery. By the 27-month point, full-time employment had recovered to its pre-recessionary level. However, the next recession in 1990 saw even slower growth, taking 34 months to regain pre-recession employment levels. Worse yet was the recovery after the 2001 recession, which took a full 40 months to recover the jobs lost since the last business cycle peak. Unfortunately, that sluggish growth is persisting far longer than it did in either of the previous two recessions. Fifty-one months after the last business cycle peak, full-time employment levels stood 7.1% (1980s), 4.2% (1990s), and 2.4% (2000s) higher than at the start of their respective recessions. If recent growth in full-time employment had matched the rate following the 1990 recession, we would have 2.1 million more people with full-time jobs today.
As for part-time work, those numbers rose steeply for about 14 months after the onset of the 1981 recession before flattening out. In the 1990s, there was a slow rise through the recession and recovery. And, in the most recent recession and recovery, there was an initial decline followed by a slow rise. Fifty-one months after the start of the recession, part-time employment was up 10.7% (1980s), 4.2% (1990s), and 5.4% (2000s) from pre-recession levels. If the current trend had followed the pattern of the 1990s recovery, there would be 280,000 fewer part-time jobs today.
Comparing just the 2001 and 1990 recessions, growth in full-time work was stronger in the previous recovery, whereas growth in part-time work was stronger in the most recent recovery. The large share of part-time jobs created in this recovery may be evidence of a "just-in-time" approach to hiring, employers' hesitancy to hire full-time workers with higher fixed benefit costs, and an overall weaker-than-usual demand for labor.
DATA NOTE: In this month's JobWatch we rely on the Current Population Survey (CPS), also known as the household survey. While the CPS enables accurate reporting on the unemployment rate, we note that it has limited reliability in total employment counts. Only the household survey, however, allows us to analyze the differential growth effects of the business cycle on full-time and part-time employment. We construct a three-month moving average to smooth month-to-month variation. Employed persons are classified in the CPS as full- or part-time workers based on their usual weekly hours at all jobs. Full-time workers are those who usually work 35 hours or more per week (at one or more jobs); part-time workers usually work less than 35 hours per week.
1. The author would like to remind readers that employment levels within the household survey should not be relied on for their month-to-month accuracy. The establishment survey is superior for those counts.
If you wish to unsubscribe yourself from this list, visit
Economic Policy Institute
Copyright ©2005 Economic Policy Institute. All rights reserved.