Although the 11th recession of the postwar period officially ended in June 2009, one economic indicator that is very visible in people's daily lives—the unemployment rate—has continued to rise. With the unemployment rate at 9.8% in November 2010, those still employed or able to find jobs are quite likely to know personally others who have been less fortunate. The still high unemployment rate partly reflects the slow pace at which employers have been adding workers to their payrolls despite enactment of job creation legislation in February 2009 (the American Recovery and Reinvestment Act, P.L. 111-5), March 2010 (the Hiring Incentives to Restore Employment Act, P.L. 111-147), and August 2010 (the Education Jobs Fund at Title I of the FAA Air Transportation Modernization and Safety Improvement Act, P.L. 111-226).
This report addresses the question of when one might reasonably expect to see sustained improvement in the unemployment rate and a steady resumption of job growth following a recession's end. It first provides an explanation of why it is unlikely that the unemployment rate would begin trending downward and the number of jobs on employer payroll would begin trending upward immediately after the start of a recovery. It then analyzes the trend in the unemployment rate and jobs data before and after the bottom of the prior 10 business cycles to confirm these statements.
Explanation of Trends in Unemployment and Jobs Around the Bottom of the Business Cycle
A few definitions are called for before proceeding with the explanation. To begin with, the official definition of the unemployment rate prescribed by the U.S. Bureau of Labor Statistics (BLS) is the percentage of persons in the civilian labor force age 16 and older who are unemployed. The labor force, in turn, is defined as the number of employed and unemployed individuals in the civilian noninstitutional population age 16 and older. Lastly, unemployment is defined as the number of people who actively searched for work in the four weeks before the week in which the Census Bureau administers the Current Population Survey to members of households. Mathematically, the unemployment rate is expressed as:
|
the number of unemployed persons |
|
|
the number of unemployed and employed persons |
|
The unemployment rate consequently may move up or down depending not just on a change in the number of workers who lack jobs (the numerator) but also on a change in the number of individuals participating in the labor force (the denominator). The size of the labor force may decrease during a recession in part because unemployed workers who give up searching for jobs out of discouragement over their reemployment prospects cease to be included. Conversely, the number of labor force participants may increase during a recession if people see signs the economy is improving and think it again worthwhile to look for work. If hiring does not keep pace with discouraged workers reentering the labor force, for example, the unemployment rate will rise in the face of otherwise good economic news. The unemployment rate thus is called a lagging economic indicator: it typically improves only after other indicators have signaled an impending recovery and actually may rise for some time after a recession's end.
Firms usually are reluctant to add workers to their payrolls until convinced that the economy will continue growing (i.e., gross domestic product, GDP, will keep increasing). Before employers are fairly certain that the increase in demand for their goods and services will continue, they are likely to boost production by restoring the hours of those currently in their employ and by having employees work overtime rather than by hiring new workers or recalling laid off workers. As a result, job growth may not occur immediately upon the economy showing signs of a turnaround.1
The source of jobs data that labor market analysts generally utilize is the Current Employment Statistics program. In its establishment survey, BLS queries nonfarm employers each month about the number of jobs (employees) on their payrolls. Employers are considered to be a more accurate source for this information compared to the members of households who respond to the Current Population Survey, from which the unemployment rate and most demographic information on workers is derived. In addition, the Business Cycle Dating Committee appears to rely chiefly on changes in both national output (GDP) and employment trends from the establishment survey to determine the official beginning and end of recessions.
Trends in Unemployment and Jobs Before and After the End of the 10 Prior Recessions
The U.S. economy experienced 10 recessions since World War II, excluding the latest, which ran from December 2007 to June 2009. These recessions varied greatly in severity, as measured by the extent of decline in GDP, and their duration.2 But, without exception, the unemployment rate kept increasing until or more often after each recession had ended. (See Table 1.)
Table 1. The Unemployment Rate at a Recession's End and at the Rate's Peak
|
Dates of Business Cycle Peak and Trough |
Unemployment Rate at Trough |
Peak Unemployment Rate |
|||||
|
Level |
Date |
||||||
|
Nov. 1948 - Oct. 1949 |
|
|
Oct. 1949 |
||||
|
July 1953 - May 1954 |
|
|
Sept. 1954 |
||||
|
Aug. 1957 - Apr. 1958 |
|
|
July 1958 |
||||
|
Apr. 1960 - Feb. 1961 |
|
|
May 1961 |
||||
|
Dec. 1969 - Nov. 1970 |
|
|
Aug. 1971 |
||||
|
Nov. 1973 - Mar. 1975 |
|
|
May 1975 |
||||
|
Jan. 1980 - July 1980 |
|
|
July 1980 |
||||
|
July 1981 - Nov. 1982 |
|
|
Nov. 1982 |
||||
|
July 1990 - Mar. 1991 |
|
|
June 1992 |
||||
|
Mar. 2001 - Nov. 2001 |
|
|
July 2003 |
||||
The unemployment rate peaked at the bottom of the business cycle in three instances: at 7.9% in October 1949, at 7.8% in July 1980, and at 10.8% in November 1982. In the remaining seven recessions, the unemployment rate did not peak until after they had ended. Four times the unemployment rate kept climbing for only a few months into the economic recovery: four months after the July 1953-May 1954 recession ended, three months after the August 1957-April 1958 recession and the April 1960-February 1961 recession ended, and two months after the November 1973-March 1975 recession ended.
In contrast, the unemployment rate did not peak until well into the economic recovery in three instances. The unemployment rate kept rising for nine months after the end of the December 1969-November 1970 recession, 15 months after the end of the July 1990-March 1991 recession, and 18 months after the end of the March-November 2001 recession.
Like the unemployment rate, the employment data generally fit the picture of no steady job growth until some months after a recession's end. As shown in Figure 1, a sustained period of job growth failed to ensue until after the end of all 10 recessions. In only one of those instances—the very short January-July 1980 recession—did the employment level immediately begin trending upward.
In the other nine cases, the number of jobs on employer payrolls fluctuated for months after the recession's end. Sustained job growth occurred within three to five months of the start of seven recoveries. In sharp contrast, steady job growth did not commence until March 1992—12 months after the July 1990-March 1991 recession ended; and not until September 2003—22 months after the March-November 2001 recession ended.
The very delayed improvement in the labor market following the 1990-1991 and 2001 recessions led to the ensuing rebounds of the economy being referred to as jobless recoveries. With employment at public and private sector employers similarly having failed to show consistent improvement for well over a year since the recession's end, many observers think the nation has been experiencing another jobless recovery. While employment at firms in the private sector of the economy started to steadily rise in January 2010, it appears that the fiscal problems of local governments in particular have dampened the pace of overall job growth.3