THE URBAN
INSTITUTE
Declining Economic
Opportunity in America
THE FUTURE OF
THE PUBLIC SECTOR
Isabel V. Sawhill and Daniel P. McMurrer
A series on
the long-term
forces affecting
U.S. social policy
A
merica has always called itself “the land
These economic trends could well have politof opportunity.” Reality has never quite
ical consequences, as those affected look for
matched the rhetoric, but a number of facsomeone to blame for their downward mobility.
tors historically have brought America progressively closer to that ideal. The continued
Causes of the Decline in
expansion of opportunities to previously
Intergenerational
excluded groups, the extension of education
to an ever-increasing share of the population,
Opportunity
and the impressive economic growth that preWhat has gone wrong? Two broad ecovailed for many years all made it easier for
nomic trends lurk behind the recent decline in
opportunity to spread broadly through the
opportunity. First, economic growth has
population.
slowed significantly since 1973, causing
In recent years, however, this record has
average earnings to stagnate. Almost
not been sustained. The period since
simultaneously, earnings inequality
the early 1970s has been marked
has increased, bringing about a
by an unprecedented decline in
dramatic reversal of the trend
Today, it is more
opportunity for many, espetoward greater equality that
difficult than it has
cially young men without
had prevailed since World
ever been for young
college degrees. Today, it is
War II. Either trend, occurworkers to surpass their
more difficult than it has
ring alone, would have had
ever been for young workparents’ standard of livtroubling consequences.
ers to surpass their parents’
The two trends together
ing, thereby achieving
standard of living, thereby
have
wrought a wrenching
the proverbial
achieving the proverbial
change
in prospects for many
American dream.
American dream.
members of the younger generThe decrease in opportunity
ation.6
is illustrated by the average earnings
Slower Economic Growth
of young men. Men born between 1938 and
Upward economic mobility and strong
1947, who were ages 25 to 34 in 1972, had
economic
growth have frequently gone hand
average incomes of $30,000 that year (in
in
hand.
Indeed,
in comparing the United
1993 dollars). Men born between 1958 and
States
with
other
countries, some scholars
1967, by contrast, who were 25 to 34 in 1992,
have
suggested
that
the faster rate of ecoaveraged about $22,000 in that year—a pre1
nomic
growth
that
prevailed
in the United
cipitous drop (see Chart 1). Similar trends
States
until
recently
is
the
primary
reason this
prevail with regard to family and household
7
2
country
enjoyed
greater
social
mobility.
incomes, although an increase in the number
Productivity
has
increased
at
an
average
of two-earner families and growth in fringe
annual rate of about 2 percent since 1870.
benefits have partially offset the effects of the
3
The 1960-1973 period saw especially strong
decline in individual wages.
growth, with productivity increasing by 3.0
Overall, young men today have lower
percent per year.8
incomes than their counterparts did in earlier
4
The good news did not last. Productivity
years. This reduction in relative well-being
slowed
to a crawl after 1973—an average of
early in the earnings cycle is likely to persist
5
1.1
percent
per year between 1973 and 1993.
or even worsen as this generation ages.
No. 1, April 1996
No. 1
THE FUTURE OF THE PUBLIC SECTOR
2
for entry-level wages, where the
Because the compensation of workers
approximately 50 percent of the
“wage premium” for college gradutends to track their productivity, real
$23,000 gap.
wage growth slowed commensurately.
ates was 77 percent in 1993, an
The increase in inequality has
What is the impact on an average
been documented and analyzed by
increase from 37 percent in 1979.12
worker of this slowdown in producMost analysts attribute the
numerous researchers, who have contivity from 3.0 percent annually to 1.1
increased wage differential for educacluded that wage inequality today is
percent? Quite a lot. In 1973, the tyption primarily to a substantial increase
higher than at any time since World
ical male high school graduate could
in the demand for more-educated
War II.10 It began to increase during
expect an average full-time entrythe 1970s, surged during the early
workers, a shift that appears to have
level wage, in 1993 dollars, of
1980s, and has since continued to
been driven by factors related to tech$20,820 (see Chart 2). If productivity
increase.11
nological change. Workers who use
had continued to grow at
computers on the job
the higher rate of the earenjoyed faster wage
lier postwar period, and if
growth than other
CHART 1
that growth had benefited
workers during the
Young Adults Have Lower Incomes than
everyone equally, entry1980s, which supports
Their Parents Did at the Same Age
(Median Income of Males, 25-34, by Birth Cohort)
level wages for a male
this view.13 Indeed, the
high school graduate by
forces
related to techThousands (1993 $)
$35
1993 would have been
nological change are
$30
over $16,000 higher
considered so powerful
Males born in:
$25
($37,603). Distributed
that they are generally
1928 - 1937
equally, the actual (slowassigned a significant
$20
1938 - 1947
er) growth in productivity
percentage of the
$15
1948 - 1957
would have increased
responsibility for the
1958 - 1967
$10
entry-level wages for
overall increase in
$5
male high school graduwage inequality in the
$0
ates by only about $5,000
United States.14
Born in:
Ages 25-34 in:
Median Income (1993 $)
from 1973 to 1993 (to
1958 - 1967
1992
$22,252
$25,912).
1948 - 1957
1982
$24,913
What Can Be
1938 - 1947
1972
$30,000
Thus, the slowdown
1928 - 1937
1962
$22,971
Done?
in productivity growth by
Source: U.S. Bureau of the Census, P60-184 (1992), p. B-33.
itself has sharply curStagnating incomes
Urban Institute, 1996
tailed the opportunity for
and increasing earnings inequality have
the average male high
Males, younger workers, and
reconfigured the economic landscape
school graduate to improve on stanthose who are not college-educated
for tens of millions of Americans—
dards of living enjoyed by previous
have been particularly affected by the
particularly younger Americans.
generations. Because of the produccombined trends. Although trends in
Government did not create these
tivity slowdown, his entry-level
inequality have been the same for
problems and they are too large and
wages are only slightly higher on
men and women, male median earndeeply rooted for the public sector to
average than those of the cohort born
ings have decreased even as female
solve alone. But it can help.
twenty years earlier.9
median earnings have increased (in
Government should target its
energies to areas in which it can be
Increasing Earnings Inequality
part because of their increased work
most effective—such as reducing the
But this is not the end of the
hours). At the same time, work expedeficit, which is a drag on economic
story. An increase in earnings
rience is more highly rewarded by
employers than in the past, so
growth, and using its leverage, leaderinequality exacerbated the effects of
younger workers without this experiship, and limited resources to encourthe decline in productivity growth.
ence have also seen a disproportionage early childhood education, local
When rising inequality is also taken
ate decrease in their wages.
school reform, more private sector
into consideration, the entry-level
Most important, less-skilled,
job training, and greater portability of
wages for a male high school graduless-educated men have experienced
benefits. It should also extend various
ate working full-time were only
particularly sharp drops in real earnkinds of help—such as retraining and
$14,518 in 1993, more than $6,000
lower than entry-level wages in 1973.
ings. In part, this is a result of a sigplacement assistance—to workers
who lose their jobs. The evidence that
This is over $23,000 less than would
nificant increase in the pay differensuch assistance increases future earnhave been expected if there had been
tial between college-educated workings or employment is mixed. But it
faster growth from which everyone
ers and others, with college graduates
(male and female) earning 58 percent
can at least cushion the impact of curhad benefited equally. Thus, the commore than high school graduates in
rent trends on those least likely to
bined costs to this group of slower
1993, compared with 38 percent more
prosper in the new economy. At the
growth and rising inequality have
in 1979. The disparity is even greater
been enormous. Each accounts for
same time, other social institutions—
No. 1
have constituted a rising proportion of
“Baby Boomers and Their Parents,”
families, schools, local communities,
compensation over time.
Journal of Human Resources, vol. 30,
and businesses—must all play a role
1995, pp. 791-806.
in preparing the next generation for a
2. See Lawrence Mishel and Jared
world in which education will
5. See, e.g., Easterlin et al. (1993),
Bernstein, The State of Working America:
who find that “higher income for a cohort
increasingly determine the success of
1994-1995, M.E. Sharpe, 1994, p. 74.
at earlier ages typically foreshadows highboth individual citizens and the nation
3. The increase in the number of
er income at later ages.” Easterlin,
as a whole.
female-headed households, however, has
Richard A., Christine M. Schaeffer, and
There is no sense in pretending
exacerbated the effects of the wage
Diane J. Macunovich, “Will the Baby
decline. See, e.g., Robert I. Lerman, “The
the solutions—public or private—
Boomers Be Less Well-Off Than Their
Impact of the Changing U.S. Family
will be cheap or financing them easy.
Parents? Income, Wealth, and Family
Structure on Child Poverty and Income
Indeed, in the present fiscally conCircumstances Over the Life Cycle in the
Inequality,” Economica (forthcoming).
strained environment, making the
United
States,”
Population
and
necessary investments
Development Review,
vol. 19, 1993, p. 503.
will require shifting more
Mishel and Bernstein
CHART 2
of society’s resources
(1994) find that, in
Downward Mobility Is the Result of both
from the support of an
Slower Growth and Rising Inequality
recent years, median
aging population to edu(Entry-Level Wages, High-School Educated Males, 1973-1993)
family incomes have not
cation, training, and other
only started at a lower
types of assistance aimed
Thousands (1993 $)
income for younger
$37,603
at preparing younger famDue to Slower
cohorts, but have also
Growth
ilies to take full advangrown more slowly than
$30
tage of today’s opportuniin the past as the mem$25,912
bers of the cohorts get
ties.
Due to Rising
older.
Government should
$20
Inequality
at a minimum avoid
$14,518
6. It should be noted
actions that make matters
that real wage growth
Faster Growth
$10
worse. Proposals to
has slowed and inequaliSlower Growth
Actual
reduce assistance to
ty has increased in other
industrialized nations as
lower-income families
$0
1973
1993
well, although the trends
(such as cuts in the
Source: Lawrence Mishel and Jared Bernstein,The State of Working America: 1994-1995 (1994), p. 147.
have been most proNote: Faster growth is defined as 3.0 percent per year (actual rate of productivity growth from
Earned Income Tax
1960-1973); slower growth is defined as 1.1 percent per year (actual rate of productivity growth from
nounced in the United
Credit) or to reduce the
1973-1993). Wages are what a full-time full-year worker would earn at average entry-level wage.
States. See, e.g., Economic
Urban Institute, 1996
taxes paid by those who
Report of the President
are doing well are espe1995, p. 172.
cially misguided in the
7. See, e.g., Daniel Heath (ed.),
face of a rising tide of inequality. To
4. Although average real earnings
America in Perspective, Houghton Mifflin
have declined compared to preceding genignore this trend is to invite a political
Co., 1986.
erations, other measures of economic stabacklash that would be both divisive
tus indicate some improvement for recent
and inimical to further growth.
8. Economic Report of the President
generations relative to predecessors. For
1996, p. 332. All productivity statistics
The New Deal worked to soften
example, “earnings per adult equivalent,”
reported here are defined as output per
the rough edges of capitalism at a
which adjusts for household size, has
hour in the nonfarm business sector,
time when the foundations of a marincreased for more recent generations.
reflect the January 1996 comprehensive
ket-oriented economy were under
This apparent improvement, however, is
revisions of the national income and prodgreat stress and subject to political
due primarily to various “demographic
uct accounts, and were computed using
challenge. In the 1990s we face a simadjustments” that have been made by
chain-type output indices. At this time,
ilar challenge. We need to forge a
individuals and have had the effect of
revised data are not available prior to
public philosophy that recognizes the
countering the impact of decreasing earn1959. Earlier (unrevised) data suggest that
ings. For example, compared to earlier
benefits of living in a dynamic, techthe rate of productivity growth that pregenerations, younger adults today are
vailed between the end of World War II
nologically advanced economy but is
more likely to remain single, marry later,
and 1959 was slightly lower than the rate
also responsive to the collateral dambetween 1960 and 1973.
and have fewer children. One study conage it inflicts on many workers and
cludes that “this economic success has
their families.
9. Because this example assumes
}
Notes
1. It should be noted that these numbers slightly underestimate the actual
change in total compensation because
they do not include fringe benefits, which
been purchased at the expense of noneconomic aspects of welfare, such as family
life, leisure, privacy, and independence.”
Richard
A.
Easterlin,
Christine
MacDonald, and Diane J. Macunovich,
“How Have American Baby Boomers
Fared?,” Journal of Population and
Economics, vol. 3, 1990, p. 287. See also
John Sabelhaus and Joyce Manchester,
THE FUTURE OF THE PUBLIC SECTOR
}
that the impact of the change in productivity is distributed evenly across the labor
force, the preceding two sentences apply
to all workers—not only male high school
graduates.
10. See, e.g., Frank Levy and
Richard J. Murnane, “U.S. Earnings
Levels and Earnings Inequality: A
3
No. 1
Review of Recent Trends and Proposed
Explanations,” Journal of Economic
Literature, vol. 30, September 1992, pp.
1333-1381; and Janet L. Norwood (ed.),
Widening Earnings Inequality: Why and
Why Now, Urban Institute, 1994.
11. Lawrence F. Katz notes that
inequality has been found “along essentially every dimension one cuts the [earnings]
data and appears to remain no matter how
finely one cuts it” (Katz, “Comments and
Discussion,” Brookings Papers on
Economic Activity, vol. 2, 1994, p. 257).
12. Mishel and Bernstein (1994),
pp. 140-147.
13. Alan B. Krueger, “How
Computers Have Changed the Wage
Structure: Evidence from Microdata,
1984-1989,” Quarterly Journal of
Economics, vol. 108, February 1993, pp.
33-60.
This series, funded in part by the
Ford Foundation, focuses on challenges for policymaking in the 21st
century.
Advisory Board
C. Eugene Steuerle
Christopher Edley, Jr.
Edward M. Gramlich
Hugh Heclo
Pamela Loprest
Demetra S. Nightingale
Isabel V. Sawhill
William Gorham
Published by
Isabel V. Sawhill is a Senior Fellow at the Urban Institute, occupying the Arjay Miller Chair in public policy. She is editor of Welfare
Reform: An Analysis of the Issues (Urban Institute, 1995).
Daniel P. McMurrer is a Research Associate at the Urban
Institute. His principal areas of research are income inequality and job
security.
Sawhill and McMurrer are currently working on a series of issue
briefs to be published over the next year on the general theme of
“Opportunity in America.”
Telephone: (202) 833-7200 n
Fax: (202) 429-0687 n
THE URBAN INSTITUTE
THE FUTURE OF THE PUBLIC SECTOR
2100 M Street, N.W.
Washington, D.C. 20037
Address Correction Requested
The Urban Institute
2100 M Street, N.W.
Washington, D.C. 20037
Copyright © 1996
The views expressed are those of
the authors and do not necessarily
reflect those of the Urban Institute,
its board, its sponsors, or other
authors in the series.
Extra copies may be requested by calling (202) 857-8687.
Designed by Robin Martell and
Barbara Willis
E-Mail: paffairs@ui.urban.org n
Web Site: http://www.urban.org
Nonprofit Org.
U.S. Postage
PAID
Permit No. 8098
Washington, D.C.