THE UNEMPLOYMENT RATE AND THE POTENTIAL SUPPLY OF LABOR

CRS Report for Congress
of Labor
August 16, 1999
Linda Levine
Specialist in Labor Economics
Domestic Social Policy Division

Congressional Research Service ˜ The Library of Congress

ABSTRACT
This report examines alternatives to the official unemployment rate to ascertain the pool of
labor potentially available to meet rising employer demand during the more than 8-year-old
economic expansion. The analysis provides one explanation for the current low-inflation
environment despite an official unemployment rate that is below the full employment rate or
non-accelerating inflation rate of unemployment. The report will not be updated.

The Unemployment Rate and the Potential Supply of Labor
Summary
Many eagerly await the release of the unemployment rate by the U.S. Bureau of
Labor Statistics (BLS) each month. The Federal Reserve Board, for one, monitors
the unemployment rate’s fluctuations for indications of the availability of labor relative
to employer demand (i.e., the tightness of the labor market) and, in turn, what such
tightness portends for the inflation rate. The unemployment rate represents the share
of persons in the labor force (i.e., the employed and unemployed) who lack jobs and
are actively seeking them. A low and falling unemployment rate, such as experienced
during the latter half of the 1990s, suggests there is a relatively small and dwindling
number of workers available to fill jobs. This has prompted some analysts to
speculate that we are running out of workers to perpetuate the nation’s now longest
peacetime expansion in the post-war period.
The unemployment rate does not tell the whole story about how much labor
remains available to meet employer demand, however. As defined by the BLS, the
unemployment rate is the fraction of the civilian noninstitutional population age 16 or
older who were not employed during the week in which the Current Population
Survey is conducted and who sought work in the prior 4-week period. This measure
understates the potential supply of labor because it captures neither the “hidden
unemployed” nor the “underemployed.” The inclusion of these two groups provides
a more complete tale of the amount of untapped labor in the population and one
explanation for the inflation rate’s failure to accelerate thus far in response to a
seeming dearth of workers.
The hidden unemployed, unlike the officially unemployed, are not in the labor
force. The notion underlying the concept of the hidden unemployed is that, although
these persons are less firmly attached to the labor market than are those officially
counted as unemployed, they too desire and are available for employment and have
searched for jobs in the not-too-distant-past; and consequently, they should be
considered part of the current supply of labor. In July 1999, there were somewhat
over 1.1 million “marginally attached persons.” If they were added to the official
count of unemployed persons in that month (slightly above 6.3 million), it would have
raised the potential supply of labor by 18% to nearly 7.5 million individuals and the
unemployment rate from 4.5% (the official, not seasonally adjusted, unemployment
rate in July 1999) to 5.2%. In addition, there were over 3.5 million employed persons
who would have preferred to supply more hours of labor to employers in July 1999.
If these underemployed or partially employed persons were included with (fully)
unemployed workers, the size of the untapped labor supply would have been 56%
larger or nearly 9.9 million. And, if marginally attached as well as underemployed
persons were added to the unemployed, the number of individuals potentially available
to meet employer demand would have been 74% larger or almost 11 million. The
inclusion of both these groups would have produced a unemployment rate of 7.7%
compared to July’s official unemployment rate of 4.5%. Just such comparisons might
have underpinned Federal Reserve Board Chairman Alan Greenspan’s observation
while testifying before the Joint Economic Committee in June 1999 that “labor market
tightness has not, as yet, put the current expansion at risk.”

Contents
The Unemployment Rate..........................................1
The Potential Supply of Labor......................................2
Marginally Attached Persons...................................2
Involuntary Part-Time Workers.................................5
Alternative Measures of Labor Utilization.............................6
Concluding Remarks.............................................7
List of Tables
Table 1. The Civilian Unemployment Rate............................2
Table 2. The Number of Marginally Attached Persons...................4
Table 3. The Number of Workers Involuntarily Employed Part-Time........6
Table 4. Alternative Measures of Labor Utilization......................7

The Unemployment Rate and the Potential
Supply of Labor
Many public and private institutions as well as individuals eagerly await the
release of the unemployment rate, usually on the first Friday of each month, by the
U.S. Bureau of Labor Statistics (BLS). The Federal Reserve Board, for one,
monitors the unemployment rate’s fluctuations for indications of the availability of
labor relative to employer demand (i.e., the tightness of the labor market) and, in turn,
what such tightness portends for the inflation rate. The Board could act to slow the
pace of economic growth if it thought too much of an imbalance in the labor market
would cause the rate of increase in prices for goods and services to accelerate.
The unemployment rate represents the share of persons in the labor force (i.e.,
the employed and unemployed) who lack jobs and are actively seeking them. A low
and falling unemployment rate, such as experienced during the latter half of the 1990s,
suggests there is a relatively small and dwindling number of workers ready to fill
already existing or new positions. This has prompted some analysts to speculate that
we are running out of workers to perpetuate the now longest peacetime expansion in
the postwar period.
The Unemployment Rate
Following the end of the 1990-1991 recession in March, the unemployment rate
took an unusual course. To begin with, it actually rose from 6.8% at the bottom of
the recession to the 7% range from October 1991 through June 1993. (See Table 1.)
The unemployment rate then fell below what many economists consider to be full
employment rate or the non-accelerating inflation rate of unemployment (NAIRU):
between September 1994 and June 1997, the unemployment rate was in the 5% range
before dropping into the 4% range where it remains to date. What has many
observers perplexed is that even at such a low level by historical standards, the
seeming scarcity of workers has not yet exerted much upward pressure on the
inflation rate.
The unemployment rate does not tell the whole story about how much labor
remains available to meet employer demand, however. As defined by the BLS, the
unemployment rate is the fraction of the civilian noninstitutional population age 16 or
older who were not employed during the week in which the Census Bureau conducts
the Current Population Survey (CPS) and who sought work in the prior 4-week
period. This measure understates the potential supply of labor because it neither
captures the “hidden unemployed” nor the “underemployed.” These two groups, in
combination with the unemployment rate, provide a more complete tale of the
potential or untapped supply of labor and one explanation for the inflation rate’s

failure to accelerate thus far in response to the seeming dearth of workers. An
analysis of these two groups appears below.
Table 1. The Civilian Unemployment Rate
1991 1992 1993 1994 1995 1996 1997 1998 1999
January 6.4 7.3 7.3 6.6 5.6 5.6 5.3 4.6 4.3
February 6.6 7.4 7.1 6.6 5.4 5.5 5.3 4.6 4.4
Mar. 6.8 7.4 7.0 6.5 5.4 5.5 5.1 4.7 4.2
April 6.7 7.4 7.1 6.4 5.8 5.5 5.0 4.3 4.3
May 6.9 7.6 7.1 6.1 5.6 5.6 4.9 4.4 4.2
June 6.9 7.8 7.0 6.1 5.6 5.3 5.0 4.5 4.3
July 6.8 7.7 6.9 6.1 5.7 5.5 4.9 4.5 4.3
August 6.9 7.6 6.8 6.0 5.7 5.1 4.9 4.5 ––
September 6.9 7.6 6.7 5.9 5.7 5.2 4.9 4.5 ––
October 7.0 7.3 6.8 5.8 5.5 5.2 4.7 4.5 ––
November 7.0 7.4 6.6 5.6 5.6 5.4 4.6 4.4 ––
December 7.3 7.4 6.5 5.5 5.6 5.4 4.7 4.3 ––
Annual 6.8 7.5 6.9 6.1 5.6 5.4 4.9 4.5 ––
average
Source: U.S. Bureau of Labor Statistics. Seasonally adjusted monthly data.
Note: Elsewhere in this report, the not seasonally adjusted monthly unemployment rate is compared
to other measures of labor utilization because seasonal adjustment factors have not yet been
developed for the other statistical series.
The Potential Supply of Labor
While testifying before the Joint Economic Committee on June 17, 1999, Federal
Reserve Board Chairman Alan Greenspan spoke of the strong demand for labor
reducing the “pool of available workers.” This pool is not limited to already active
jobseekers, but also includes individuals who would like to work but haven’t recently
looked for jobs and individuals who would like to work longer hours than they are
presently.
Marginally Attached Persons
What distinguishes the so-called hidden unemployed from the officially
unemployed group is that the former are not in the labor force. The notion

underlying the concept of hidden unemployment is that, although these persons are
less firmly attached to the labor market than are persons officially counted as
unemployed, they too desire and are available for employment and have searched for
jobs in the not-too-distant past; and consequently, they should be considered part of
the current supply of labor.
Since 1967, the BLS has asked CPS respondents who are not in the labor force
about whether they want a job. With the major revision of the CPS in 1994, the BLS¹
definition of individuals outside the labor force who would like to work has changed.
Currently, the hidden unemployed or “marginally attached” group is composed of
individuals in the civilian noninstitutional population age 16 or older who: (1) desire
jobs, (2) are available to work and (3) have engaged in job search in the past 12-
month period. As noted above, to be considered unemployed an individual must have
searched for work in the 4-week period preceding the survey week.
From 7% to 10% (or 4.8 to 6.2 million) of those who were not in the labor force²
between 1994 and 1998 also reported that they wanted a job. A much smaller
number who indicated that they desired employment met the two other criteria for
classification as marginally attached persons (1.1-1.8 million). Although the hidden
unemployed are a small group when compared to the official count of unemployed
persons — 1.1 million and 6.3 million, respectively, in July 1999 — their addition to
the unemployed members of the labor force raised the untapped pool of workers by

18% to almost 7.5 million individuals in that month.

People with a marginal attachment to the labor force offer various reasons for
failing to have recently sought employment. The BLS categorizes these as economic
reasons (e.g., thinks no work is available, could not find work, lacks schooling or
training, and discrimination including employer thinks person is too old) and as
noneconomic reasons (e.g., child-care and transportation problems, family
responsibilities, attending school or training, and poor health). Persons within the
marginally attached group who offer economic reasons typically are referred to as
“discouraged workers.” In July 1999, they totaled 290,000 or about one-fourth of
the larger group.
It typically was assumed that discouraged workers would be more likely than
other marginally attached persons to enter the labor force in a full-employment
economy such as currently is being experienced. According to a recent analysis,
however, discouraged workers in 1994 exhibited weaker labor force attachment and
jobholding in 1995 than other marginally attached persons — perhaps because the
former’s reasons for nonparticipation and non-employment were more intransigent.³

Castillo, Monica D. Persons Outside the Labor Force Who Want a Job. Monthly Labor¹
Review, July 1998. (Hereafter cited as Castillo, Persons Outside the Labor Force Who Want
a Job.)
Unless otherwise indicated, all data in this report are available from the BLS at²
[http://stats.bls.gov/proghome.htm].
Castillo, Persons Outside the Labor Force Who Want a Job.³

As shown in Table 2, marginally attached individuals numbered slightly over 1.3
million on average in 1998 and they shrank further through the first 7 months of 1999
to somewhat more than 1.1 million. Presumably, their ranks have dwindled because
favorable economic conditions enticed them to more actively seek jobs and thereby
enable the labor force to continue to grow. Between July 1994 and July 1999, the
number of marginally attached individuals declined by 711,000. Over the same
period, the size of the labor force swelled by 8.3 million. At most then, marginally⁴
attached persons might have allowed the labor force to be one-tenth larger than it
otherwise would have been over the past 5 years.
Table 2. The Number of Marginally Attached Persons
(numbers in thousands)
1994 1995 1996 1997 1998 1999
January 2,120 1,783 1,737 1,615 1,479 1,358
February 1,951 1,721 1,838 1,546 1,478 1,279
March 1,832 1,732 1,584 1,471 1,426 1,245
April 1,770 1,390 1,516 1,480 1,278 1,257
May 1,659 1,504 1,475 1,431 1,213 1,148
June 1,777 1,574 1,684 1,428 1,213 1,228
July 1,844 1,568 1,490 1,281 1,328 1,133
August 1,726 1,510 1,436 1,298 1,251 ––
September 1,858 1,583 1,518 1,363 1,377 ––
October 1,663 1,587 1,447 1,284 1,242 ––
November 1,674 1,542 1,503 1,337 1,240 ––
December 1,810 1,619 1,463 1,453 1,196 ––
Annual 1,807 1,593 1,558 1,416 1,310 ––
average
Source: U.S. Bureau of Labor Statistics. Monthly data are not seasonally adjusted.

Monthly data for the marginally attached group are not available on a seasonally adjusted⁴
basis. It is expected that seasonal adjustment will come later when BLS has more experience
with the new series. Until such time, comparisons should be made for the same month year-
over-year rather than within the same year month-to-month. When monthly data for the
marginally attached group are used in conjunction with other labor force series not seasonally
adjusted data for the latter should be used as well.

Involuntary Part-Time Workers
The vast majority of part-time workers — more than four out of five — want to
be employed fewer than 35 hours per week. They are referred to as voluntary part-
timers. The remaining workers employed between 1 and 34 hours a week want to
work longer hours. They are referred to as involuntary part-timers, the partially
employed or underemployed, or persons employed part-time for economic reasons.⁵
Although this group has increased as a share of total part-timers in recent decades,
cyclically induced fluctuations aside, involuntary part-time workers remain a small
share of the labor force. In 1998, for example, persons employed part-time for
economic reasons comprised less than 3% of the civilian labor force.⁶
Just as the unemployment rate continued to rise for some time after the end of
the recession in March 1991, so too did this measure. The number of persons
involuntarily employed part-time did not start to decrease until 1994, although some
of the decline from 6.5 million in 1993 to 4.6 million in 1994 is due to the revision of
the CPS that began in that year.⁷
The extent of underemployment has continued to shrink since 1994 with the rate
of decrease picking up substantially in the late 1990s. (See Table 3.) In 1995, the
number of workers involuntarily employed part-time dropped by 3.3% and in 1996,
by 3.5%. In 1997, the pace quickened to 5.7% and in 1998, the extent of
underemployment dropped by 9.9%. This trend likely is related to the initially weak
recovery in the labor market from the 1990-1991 recession and more rapid job growth
later in the decade.
Most recently, there were over 3.5 million workers who would have preferred
to supply more hours of labor to employers. If these underemployed workers were
included with (fully) unemployed workers, the size of the potential labor supply would
have been 56% larger or nearly 9.9 million rather than 6.3 million workers. And, if
marginally attached as well as underemployed persons (almost 4.7 million) were
added to the unemployed, the amount of untapped labor available to meet employer
demand would have been 74% larger or almost 11 million individuals.

Another definition of underemployment includes persons working in jobs for which they are⁵
over qualified.
For additional information on part-time workers, see: CRS Report 98-695, Part-Time Job⁶
Growth and the Labor Market Effects of Policy Responses: An Overview, by Linda Levine.
Before 1994, people’s desire and availability for full-time jobs were inferred. From 1994⁷
onward, CPS respondents have been asked explicitly about their desire and availability for
full-time work: “[T]he reduced number of involuntary part-timers results almost entirely from
the direct question about desire for full-time work; the question on availability has little
affect.” Nardone, Thomas. Part-Time Employment: Reasons, Demographics, and Trends.
Journal of Labor Research, v. 16, no. 3, summer 1995. p. 289.

Table 3. The Number of Workers Involuntarily Employed Part-Time
(numbers in thousands)
1994 1995 1996 1997 1998 1999
January 5,235 4,848 4,320 4,541 4,299 3,815
February 4,857 4,567 4,597 4,419 4,042 3,594
March 4,987 4,566 4,569 4,277 4,011 3,703
April 4,538 4,245 4,299 4,244 3,649 3,316
May 4,649 4,351 4,175 3,891 3,602 3,281
June 5,063 4,740 4,577 4,258 4,033 3,641
July 4,841 4,749 4,646 4,279 4,025 3,537
August 4,417 4,553 4,407 4,036 3,508 ––
September 4,017 4,217 4,012 3,638 3,112 ––
October 4,132 4,092 3,973 3,602 3,086 ––
November 4,368 4,335 3,860 3,768 3,159 ––
December 4,408 4,410 4,352 3,869 3,455 ––
Annual 4,625 4,473 4,315 4,068 3,665 ––
average
Source: U.S. Bureau of Labor Statistics. Monthly data are not seasonally adjusted.
Alternative Measures of Labor Utilization
Since the mid-1970s, the BLS has calculated and regularly published alternatives
to the official unemployment rate because it is used for different purposes (e.g., as a
gauge of current economic conditions, a measure of economic hardship and as an
indicator of unused or underutilized human resources) and therefore, “may not
perfectly suit the needs or interests of all people.” These measures range from⁸
narrower (e.g., job losers and workers who have completed temporary jobs as a
percent of the civilian labor force) to broader (e.g., the unemployed plus discouraged
workers as a percent of the civilian labor force plus discouraged workers) than the
official unemployment rate (i.e., the unemployed as a percent of the civilian labor
force, U-3). The two broadest alternatives, which relate to marginally attached
persons and underemployed persons, are discussed below.

Bregger, John E., and Steven E. Haugen. BLS Introduces New Range of Alternative⁸
Unemployment Measures. Monthly Labor Review, October 1995. p. 20.

U-5 is defined as unemployed and all marginally attached persons as a percent
of the civilian labor force and all marginally attached persons. (Because marginally
attached workers are added to the numerator, they also must be added to the
denominator.) U-6 is more expansive, adding involuntary part-time workers to the
mix. As shown in Table 4, both these measures suggest that there is more labor
available to meet employer demand than revealed by the official unemployment rate.
Table 4. Alternative Measures of Labor Utilization
(percentages)
Measures 1994 1995 1996 1997 1998 July1999
U-3 6.1 5.6 5.4 4.9 4.5 4.5^a
U-5 7.4 6.7 6.5 5.9 5.4 5.2^b
U-6 10.9 10.1 9.7 8.9 8.0 7.7^c
Source: U.S. Bureau of Labor Statistics. Monthly data are not seasonally adjusted.
The official unemployment rate, i.e., the number of unemployed persons as a share of the civilian^a
labor force.
The second broadest alternative measure of labor utilization, i.e., the number of unemployed and^b
all marginally attached persons as a share of the civilian labor force and all marginally attached
persons.
The broadest alternative measure of labor utilization, i.e., the number of unemployed, marginally^c
attached and underemployed persons as a share of the civilian labor force and marginally attached
persons.
Concluding Remarks
The preceding analysis implies that we are not yet close to running out of labor
to fill newly created and already existing jobs in the still expanding U.S. economy. By
adding to the number of unemployed workers those persons who indicated that they
are marginally attached to the labor force and those employed individuals who
reported that they preferred to work more hours than they are currently, the untapped
pool of workers in July 1999 was 10,989,000 rather than 6,319,000. Based on these
additions, the unemployment rate might be more nearly 7.7% than 4.5% (the official
unemployment rate in July 1999). Just such a comparison might have underpinned
Chairman Greenspan’s observation while testifying before the Joint Economic
Committee in June of this year that “labor market tightness has not, as yet, put the
current expansion at risk.” Rather than a shortage of labor per se, what might well
produce more rapid wage and price increases in the coming months is a mismatch
between the skills possessed by available workers and the skills demanded by
employers for the available jobs.