View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

FABSF

WEEKLY LETTER

Number 95-29, September 8, 1995

Unemployment
In both policy debates and the popular press,
discussions of the employment situation in the
U.S. tend to focus heavily on the overall civilian
unemployment rate. But the overall unemployment rate is just one aspect of the employment
situation, and a focus on this one number can
conceal interesting and important changes in
the employment outlook for different groups
in the labor force. This Weekly Letter discusses
some of the characteristics of u.S. unemployment
that are important for the appropriate design of
labor market policies and for the design of monetary policies aimed at controlling inflation.
The composition of unemployment
Measurement of the unemployment rate is based
on a monthly survey of 60,000 households conducted by the Bureau of Labor Statistics. The
overall unemployment rate equals the number of
individuals in a given month who are looking for
work as a fraction of the total number of individualsin the labor force. Individuals who are not
actively looking for work are not counted as unemployed or as part of the labor force; individuals working part-time, but who might like to
work full-time, are counted as employed.
Individuals differ in many ways,and different
groups experience very different levels of unemployment. By providing an aggregate measure,
the overall unemployment rate does not answer
questions about who the unemployed are: Are
ethnic minorities more likely to be unemployed
than white members of the labor force? Are
women more likely to be unemployed than men?

Figure 1
White and Nonwhite (16 yrs. +)
Unemployment Rates

Percent

20

.,

,~

.:':

.'

...

Nonwhite

15

5

O;-'-,.,-rn"TT'TT"rnrrT,.,.,.,-rn"TT'TT"rnrrT.,..,

60

65

70

75

80

85

90

80

85

90

Figure 2
Male and Female (20 yrs. +)
Unemployment Rates

Percent

12

9
To address these questions, Figures 1 and 2 give
breakdowns of unemployment rates by race and
gender. Figure 1 shows the total unemployment
rate, together with the unemployment- rates for
whites and nonwhites. Data are monthly and
cover the period from January 1960 to June 1995.
The important role of business cycles in producing common patterns in all three rates is evident
-a rising tide does raise all boats when it comes
to unemployment-but the nonwhite rate of unemployment is roughly twice that for whites. This
2: 1 ratio holds as a rough approximation over the

6

3

60

65

70

75

FABSF

Figure 3

Unemployment Duration
Weeks

entire 35~year period shown. The vast changes in
social and legal structures in the
that have
affected the economic opportunities of nonwhites have failed to make a dent in this 2:1
relationship.

u.s.

Figure 2 shows the unemployment rates for males
and females over 20 years old. In contrast to the
data on race differences in Figure 1, gender differences in unemployment rates have changed
significantly over the past 35 years. During the
1960s and 1970s, the female unemployment rate
averaged over 30 percent higher than the male
rate. But labor market trends during the 1980s
and 1990s have seen this gap narrow and even
reverse. Unfortunately, the convergence of male
and female unemployment rates did not occur as
a result of the female rate falling to the lower levels previously experienced by males. Instead, as
Figure 2 shows, the male rate rose, both absolutely and relative to the female rate, during the
1980 recession, and has remained roughly equal
to the female rate ever since. In fact, during the
first half of the 1990s, the unemployment rate for
females over 20 years of age has averaged less
than that for the corresponding group of males
(5.5 percent versus 5.9 percent).

The duration of unemployment
Unemployment rates provide one measure of
labor market conditions, and another is the
duration of unemployment, that is, how long an
individual remains unemployed. The average
(mean) duration of unemployment during the
first six months of 1995 was 16.9 weeks.
Duration is a useful measure of labor market
conditions. Because the overall unemployment
rate represents an average of the number of individuals who experience some unemployment,
weighted by the length of time they are unemployed, the same overall unemployment rate
could occur if many people experience short
spells of unemployment or if a much smaller
number experience very tong spells of unemployment. Job training and placement policies might
have little impact on unemployment if unemployment spells are already quite short, while such
programs might have a major impact if unemployment arises predominately from long spells
of joblessness.
The solid line in Figure 3 shows mean duration
since 1960. Duration clearly fluctuates with the
business cycle: During recessions, total unemployment rises and the average length of a spell
of unemployment increases, and in expansions,

25

20

rv;

Mean

15

10

......

:-

~

....~....~....:
.............:/.

5
Median

60

65

70

75

80

85

90

the opposite occurs. Since 1969, average duration has trended upwards. While unemployment
for the first six months of 1995 averaged 5.6 percent, the same as the average rate for 1972, mean
duration in 1995 has averaged almost 17 weeks
while it was only 12 weeks in 1972. In 1972, almost half (46 percent) of the unemployed had
been without jobs for under five weeks; in 1994,
this figure was only about a third (34 percent). In
contrast, the fraction of people who had been unemployed for more than 27 weeks was 16 percent
in 1972 and 20 percent in 1994. Thus, the same
overall unemployment rate today corresponds to
more long-term unemployment than it did earlier.
The dotted line in Figure 3 shows median duration since 1967, which, many have argued, is a
more appropriate focus than average duration of
unemployment. By definition, half of all unemployment spells will be shorter than the median
duration and half will be longer. During the first
half of 1995, the median duration was 8.1 weeks;
in other words, half of all completed spells of unemployment were less than 8.1 weeks in length.
If a small fraction of the unemployed experience
very long periods of unemployment while most
unemployed are unemployed for only short periods, the mean duration will exceed the median
duration. Median duration is, in fact, significantly less than mean duration in the U.S. In
1994, for example, mean duration was almost
nineteen weeks while the median duration was
only nine weeks and one third of all spells of unemployment lasted less than five weeks. Thus,
U.S. unemployment tends to consist of a large
number of relatively short spells of unemploy-

ment combined with a smaller number of very
long spells of unemployment. However, a spell of
unemployment may not end because the individual has found a job. Analyzing data from 1974,
Clark and Summers (1979) concluded that almost
half of all spells of unemployment ended because
the unemployed worker withdrew from the labor
force.
Comparing Figures 1 and 3 reveals some important developments in U.s. labor markets in the
1990s. The overall unemployment rate generally
has declined during the last three years. Earlier
experience would have suggested that both the
mean and median duration of unemployment
would have followed a similar path. But Figure 3
shows that after rising sharply during the 1990
recession, the duration measures remained high
after 1992 even though the overall unemployment rate was falling. This suggests the growing
importance of long-term unemployment in the
U.S.

The natural rate
Labor market conditions are one factor considered by the Fed in deciding on the appropriate
stance of monetary policy. While popular discussions tend to focus on the overall unemployment
rate, it is the relationship between the actual rate
and the economy's so-called natural rate that
most economists believe is critical for judging
short-term changes in inflationary pressures. The
natural rate is the long-term average unemployment rate, and it depends on the microeconomic
structure of the labor market, the supply of workers and their skills, and the economy's production technology. One recent estimate places the
natural rate at about 61;4 percent (Weiner
1993).

u.s.

The natural rate plays an important role in the
conduct of monetary policy, since sustained periods with unemployment below the natural rate are
typically associated with increases in the rate of
inflation. Thus, comparisons between the actual
overall unemployment rate and the estimated
natural rate can provide an indication of the likelihood of inflationary pressures in the economy.
As unemployment falls below the natural rate, labor markets tighten, and upward pressures on

\'vages and prices build.
The inflationary effects of economic expansions
will depend, in part, on the responsiveness of

wages to tightening labor markets. And the degree of tightness in labor markets associated with
a given overall unemployment rate may vary over
time. For example, if the long-term unemployed
exert less downward pressure on wages and
prices, perhaps due to more marginal attachment
to the labor force or to concentration in low-wage
or minimum-wage occupations, changes in the
distribution of unemployment between shortterm and long-term unemployment can affect inflationary pressures. If short-term unemployment
falls while long-term unemployment rises, there
might be increased pressures on wages even
though the overall unemployment rate may have
remained unchanged. Thus, changes in such factors as the distribution of unemployment between
long-term and short-term spells can provide signalsabout potential shifts in the natural rate that
can usefully supplement the information contained in the actual rate of unemployment.

Conclusions
Even though overall unemployment in the U.S. is
now at levels similar to those of the early 1970s,
just prior to the first oil price shock, the composi. tion of unemployment has changed. Unemployment among males is much higher, for example.
And the duration of unemployment, whether
measured as the mean or the median, has risen
dramatically. Long-term unemployment appears
to be a more serious problem than it was in the
early 1970s, even though average unemployment
is roughly the same. One aspect of
unemployment that has not changed is the ratio of
nonwhite to white unemployment; nonwhites
continue to have rates of unemployment that are
roughly twice the level experienced by whites.

u.s.

Carl E. Walsh
UC Santa Cruz
Visiting Scholar, FRBSF

References
Clark, K.B., and L.H. Summers. 1979. "Labor Market
Dynamics and Unemployment: A Reconsideration!' Brookings Papers on Economic Activity 1, pp.

13-60.
Weiner, S. 1993. "New Estimates of the Natural Rate
of Unemployment." FRB Kansas City Economic Review 4, pp. 53-69.

Opinions expressed in this newsletter do not necessarily reflect the views of the management of the Federal Reserve Bank of
San Francisco, or of the Board of Governors of the Federal Reserve System. Editorial comments may be addressed to the editor
or to the author. Free copies of Federal Reserve publications can be obtained from the Public Information Department, Federal
Reserve Bank of San Francisco, P.O. Box 7702, San Francisco 94120. Phone (415) 974-2246, Fax (415) 974-3341. Weekly Letter
texts and other FRBSF publications and data are available on FedWest Online, a public bulletin board service reached by setting
your modem to dial (415) 896-0272.

Research Department

Federal Reserve
Bank of
San Francisco
P.O. Box 7702
San Francisco, CA 94120

Pnnted on recycled paper Q
with soybean Inks.
W

~

~

Index to Recent Issues of FRBSF Weekly Letter

DATE

NUMBER TITLE

2/10
2/17
2/24
3/3
3/10
3/17
3/24
3/31
4/7
4/14
4/21
4/28
5/5
5/12
5/19
5/26
6/9
6/23
7/7
7/28
8/4
8/18
9/1

95-06
95-07
95-08
95-09
95-10
95-11
95-12
95-13
95-14
95-15
95-16
95-17
95-18
95-19
95-20
95-21
95-22
95-23
95-24
95-25
95-26
95-27
95-28

Central Bank Credibility and Disinflation in New Zealand
Western Update
Reduced Deposit Insurance Risk
Rules vs. Discretion in New Zealand Monetary Policy
Mexico and the Peso
Regional Effects of the Peso Devaluation
1995 District Agricultural Outlook
Has the Fed Gotten Tougher on Inflation?
Responses to Capital Inflows in Malaysia and Thailand
Financial Liberalization and Economic Development
Central Bank Independence and Inflation
Western Banks and Derivatives
Monetary Policy in a Changing Financial Environment
Inflation Goals and Credibility
The Economics of Merging Commercial and Investment Banking
Financial Fragility and the Lender of Last Resort
Understanding Trends in Foreign Exchange Rates
Federal Reserve Policy and the Predictability of Interest Rates
New Measures of Output and Inflation
Rebound in U.s. Banks' Foreign Lending
Is State and Local Competition for Firms Harmful?
Productivity and Labor Costs in Newly Industrializing Countries
Using Consumption to Track Movements in Trend GOP

AUTHOR
Hutchison
Mattey/Dean
Levonian/Furlong
Spiegel
Moreno
Mattey
Dean
Judd/Trehan
Glick/Moreno
Huh
Parry
Laderman
Glick/Trehan
Judd
Kwan
Schaan/Cogiey
Kasa
Rudebusch
Motley
Zimmerman
Mattey/Spiegel
Golub
Cogley/Schaan

The FRBSF Weekly Letter appears on an abbreviated schedule in June, July, August, and December.