ArticlePDF Available
Upjohn Institute Press Book Chapters Upjohn Research home page
Removing Barriers to Work for Older Americans
Katharine G. Abraham
University of Maryland
Susan N. Houseman
W.E. Upjohn Institute
This title is brought to you by the Upjohn Institute. For more information, please contact
Abraham, Katharine G., Susan N. Houseman. 2008. "Removing Barriers to Work for Older Americans." In A Future of Good Jobs? :
America's Challenge in the Global Economy, Timothy J. Bartik, Susan N. Houseman, editors. Kalamazoo, MI: W.E. Upjohn Institute,
pp. 161-202.
Removing Barriers to
Work for Older Americans
Katharine G. Abraham
University of Maryland and NBER
Susan N. Houseman
W.E. Upjohn Institute
Over the next dozen years, as the baby boomers age, the share of the
population aged 55 and older is projected to grow dramatically, from
21.4 percent in 2000 to 25.1 percent by 2010 and 29.5 percent by 2020.
Over the same period, the share of the population aged 25–54—the
age group that historically has been attached most strongly to the labor
market—is projected to fall from 43.4 percent in 2000 to 40.8 percent
in 2010 and 37.7 percent in 2020 (U.S. Census Bureau 2002). Among
other anticipated consequences, the aging of the population threatens
the solvency of both the Social Security and the Medicare systems. Re-
ecting concerns over these trends, there is a growing consensus that
increased employment among older Americans would be in the public
Congress already has taken several important steps to encourage
work at older ages. Recent changes to the law allow Social Security
bene ciaries to earn more money without having their bene ts reduced
and permit workers under the normal retirement age to phase into retire-
ment by collecting pension bene ts while working a reduced schedule.
Congress also has raised the age at which workers may collect full So-
cial Security bene ts—in effect making bene ts less generous—and
this too can be expected to increase older Americans’ desire to work.
Developments in the private sector, most notably the shift from de ned
bene t retirement plans to less generous de ned contribution plans and
the shrinking share of employers who offer retiree health bene ts, have
reinforced the effects of public policy changes. Americans are not sav-
tb08fogjch5.indd 161tb08fogjch5.indd 161 9/10/2008 12:47:02 PM9/10/2008 12:47:02 PM
162 Abraham and Houseman
ing enough to compensate for the reduction in public and private retire-
ment bene ts. If they wish to maintain their standard of living, many
Americans will need to work later in life.
Although recent changes to federal policy have altered nancial
incentives in ways that should make it more attractive for older Ameri-
cans to work, these policies have not addressed the barriers to nding
suitable employment that older workers frequently face. Survey evi-
dence shows that a high percentage of older Americans already wish to
remain employed rather than withdrawing fully from the labor market,
but many want or need to work fewer hours or to nd less physically de-
manding jobs. Our research indicates that the need to make a job transi-
tion, particularly a job transition that involves a change of employer, is
a major impediment to continued employment for seniors. In addition,
as job stability has declined and the incidence of worker dislocation has
risen over the last 20 years, a growing number of older Americans have
found themselves involuntarily out of work and searching for a new job
late in life. Many are unsuccessful in their search.
Although labor force participation among older women has grown
in the last two decades, it has declined among men in their fties. The
decline in employment among older, less-educated men has been pre-
cipitous: labor force participation among men in their fties with less
than a high school education fell by 11 percentage points over the last
two decades. The current low rate of labor force participation among
men in their fties—traditionally a group considered to be of preretire-
ment age—does not portend well for increasing the labor force partici-
pation of these cohorts as they age.
Why is it so dif cult for many older Americans, particularly the less
educated, to transition to new employment? Part of the explanation may
be that older workers—especially those who have worked for a single
employer for an extended period and thus have no recent experience
with having to nd a job—do not have a clear idea about how to search
for employment. Employer reluctance to hire older workers is another
factor. Unwarranted stereotyping accounts for some of this reluctance
to hire older workers, but more legitimate concerns about older work-
ers as potential employees play a role as well. Lack of technical skills,
low perceived returns to training, and high health insurance costs are
among the most common reservations that employers cite about hiring
older workers.
tb08fogjch5.indd 162tb08fogjch5.indd 162 9/10/2008 12:47:07 PM9/10/2008 12:47:07 PM
Removing Barriers to Work for Older Americans 163
Job training and employment programs, currently funded under the
Workforce Investment Act (WIA) and, to a lesser degree, the Trade Ad-
justment Assistance Act (TAA) and the Older Americans Act (OAA),
are the primary active labor market programs through which the federal
government seeks to overcome impediments to employment faced by
workers. In real terms, federal funding for these programs has fallen
signi cantly over the last decade. Moreover, services to older work-
ers have been deemphasized under WIA compared to what they were
under the Job Training Partnership Act (JTPA), WIAs predecessor, de-
spite the fact that the population is aging and older workers’ needs for
such services likely have risen. Some states and private organizations
have begun to take steps to address the impediments to employment
that older workers face. Many of these initiatives have focused on pub-
lic relations efforts to reduce what program managers perceive to be
widespread discrimination by employers against older workers. Some
state and private initiatives also have sought to improve the delivery of
employment and training services to older job seekers, but funding for
these initiatives has been limited.
Failure to develop and implement effective programs to retrain
older workers and place them in jobs has high public costs. Among
other concerns is the fact that many of those who fail to nd work end
up on public assistance in the form of Social Security Disability Insur-
ance (SSDI), at least in some cases not because they cannot work but
because they are unable to nd work. Partly as a result, the costs of the
SSDI program are soaring. Although the dif culties that older workers
experience as they seek to transition to new jobs will not be easy to
overcome, we propose the following ve policy steps in order to begin
to address the problem seriously:
1) Increase funding for employment and training programs that
serve older workers.
2) Modify performance standards for WIA service providers to
eliminate disincentives to serve older workers.
3) Experiment with promising approaches to serving an aging
workforce more effectively, including
Improving outreach to seniors and to their potential employ-
tb08fogjch5.indd 163tb08fogjch5.indd 163 9/10/2008 12:47:07 PM9/10/2008 12:47:07 PM
164 Abraham and Houseman
Posting older-worker specialists who are knowledgeable
about the employment and training issues seniors often con-
front at “one-stop” centers, and
Providing specialized technical skills training for seniors.
4) Evaluate promising initiatives using rigorous methodologies
to determine whether and to what extent they improve older
workers’ employment prospects and would be cost-effective if
adopted nationally.
5) Reform health care nancing to reduce disincentives to hiring
older workers.
We elaborate upon these proposals later in the chapter.
Various researchers have noted the increasing rate of labor force
participation among older Americans in recent decades (Burtless and
Quinn 2002; Munnell and Sass 2007; Purcell 2005; Quinn 1999). Al-
though the share of men aged 55 and older who were active in the la-
bor force fell from the early 1900s through the mid-1980s, labor force
participation rates leveled off beginning in 1985 and have risen slightly
since the mid-1990s. Among women, the pre-1985 trend towards earlier
retirement was offset by rising labor force participation overall, with
the result that labor force participation rates among women 55 and older
were relatively at from the mid-1960s through the mid-1980s. Since
the mid-1980s, labor force participation among older women aged 55
and older has trended upwards (Federal Interagency Forum on Aging-
Related Statistics 2006).
These aggregate trends mask important differences, however, in
trends across education groups and across more re ned age categories.
Although labor force participation rates have increased since the 1990s
among older women in all age groupings (Figure 5.1), the increase in
male labor force participation in recent years has occurred only among
men over age 60 (Figure 5.2B) and has been most pronounced among
men over age 65 (Figure 5.2C). As shown in Figure 5.2A, labor force
tb08fogjch5.indd 164tb08fogjch5.indd 164 9/10/2008 12:47:07 PM9/10/2008 12:47:07 PM
Removing Barriers to Work for Older Americans 165
participation among men in their fties—men traditionally considered
to be of preretirement age—has actually continued to fall.
One possible explanation for the decline in labor force participa-
tion among men in their fties could be that their nancial situation has
improved, either because their lifetime earnings have risen or because
their wives are now more likely to be employed, making earlier retire-
ment a more viable option. Examination of labor force participation
trends by education level suggests that this is not what has happened.
The labor force participation rate for college-educated men in their f-
ties has been fairly stable over the last 20 years but has declined for
those with lower levels of education, especially those with less than a
high school education. Between 1984 and 2005, the labor force partici-
pation rate for men aged 50–59 with less than a high school education
dropped by 11 percentage points, and the decline for men aged 50–54
was almost as large as that for men 55–59. This drop can be explained
neither by rising income levels (real wages for less-educated men have
fallen) nor by increased spousal employment (labor force participation
rates for women with less than a high school education have been stag-
nant). Below we discuss some of the factors that underlie these trends
and the special policy challenges they pose.
For the past few decades, the health of older Americans has been
improving, while changes in the mix of occupations associated with the
growth of the service economy, as well as technological advances affect-
ing many blue-collar jobs, have made work less physically demanding.
In addition, life expectancies have increased steadily, increasing in turn
the nancial resources required to maintain individuals’ standard of liv-
ing over their lifespan (Munnell and Sass 2007; Technical Panel on As-
sumptions and Methods 2003). In the past, rising incomes and generous
public and private retirement bene ts made it attractive for Americans
to retire at younger ages despite their increased life expectancy (Costa
1998). In the future, however, nancial incentives are likely to make it
more attractive for Americans to keep working longer.
tb08fogjch5.indd 165tb08fogjch5.indd 165 9/10/2008 12:47:08 PM9/10/2008 12:47:08 PM
C. Women 65–69
Figure 5.1 Trends in Labor Force Participation Rates of Older Americans, 1984–2005, by Gender and Age (Women)
A. Women 50–59 B. Women 60–64
tb08fogjch5.indd 166tb08fogjch5.indd 166 9/10/2008 12:47:08 PM9/10/2008 12:47:08 PM
C. Men 65–69
Figure 5.2 Trends in Labor Force Participation Rates of Older Americans, 1984–2005, by Gender and Age (Men)
A. Men 50–59 B. Men 60–64
tb08fogjch5.indd 167tb08fogjch5.indd 167 9/10/2008 12:47:09 PM9/10/2008 12:47:09 PM
168 Abraham and Houseman
Social Security Reforms
Several features of the Social Security system discouraged work
at older ages in the past. Recent changes have greatly reduced if not
eliminated these disincentives. First, the Social Security earnings test,
which determines any reduction in current monthly bene ts for those
with earnings from employment, has been liberalized for those between
age 62 and the normal retirement age and eliminated for those above
the normal retirement age. These changes allow bene ciaries to earn
more without experiencing a reduction in their current Social Security
bene ts. Although under previous law those whose bene ts were re-
duced because of the earnings test could expect to receive higher future
bene ts, this fact seems to have been poorly understood by bene t re-
cipients, and consequently the liberalization of the earnings test appears
to be one of the factors responsible for the recent increase in labor force
participation at older ages (Munnell and Sass 2007).
Second, new rules regarding the delayed retirement credit are being
phased in. When fully implemented in 2008, these rules will approxi-
mately equate the actuarial present value of Social Security bene ts
received by those who choose to delay receipt of bene ts to the value
for those who begin collecting bene ts at the normal retirement age. By
reducing the penalty previously imposed on those who chose to con-
tinue working past the normal retirement age, the delayed retirement
credit appears already to have increased work among those age 65 and
older (Munnell and Sass, 2007).
A third change whose full effect has not yet been felt is the sched-
uled increase in the normal retirement age—the age at which individu-
als may begin to receive full bene ts—from 65 to 67. Although workers
still may retire and begin to collect Social Security bene ts at 62, their
monthly bene t amount will be proportionately lower than would have
been true in the past, re ecting the larger actuarial reduction needed
to equate the expected present value of lifetime bene ts for a person
retiring at age 62 to that for someone retiring at age 67 as opposed to
age 65. This is a reduction in bene ts that should increase the number
of Americans who want to work at older ages (Munnell and Sass 2007;
Thompson 2004).
tb08fogjch5.indd 168tb08fogjch5.indd 168 9/10/2008 12:47:09 PM9/10/2008 12:47:09 PM
Removing Barriers to Work for Older Americans 169
Changes to Private Sector Retirement Plans
Roughly half of workers of retirement age are covered by an em-
ployer pension plan, but the character of those plans has changed a great
deal. Data from the Survey of Consumer Finance show that the share of
those with a pension covered by a de ned bene t plan fell from 87 per-
cent to 44 percent between 1983 and 1998; over the same time period,
the share covered by a de ned contribution plan rose from 40 percent
to 79 percent (Friedberg and Webb 2005). Analysis of data collected
from employers by the Bureau of Labor Statistics (BLS) shows that
the shift away from de ned bene t plans towards de ned contribution
plans continued through 2005. In addition, many employers offering
de ned bene t plans have converted them to cash balance or pension
equity plans that in important respects are more like a de ned contribu-
tion plan than the traditional de ned bene t plan (Costo 2006).
The ongoing shift in employer-sponsored retirement plans from
de ned bene t to de ned contribution plans is providing additional -
nancial incentives to workers to retire at an older age. The traditional
de ned bene t plan imposes a signi cant nancial penalty for working
past a certain age; in contrast, the present value of retirement bene ts
under a de ned contribution plan continues to grow so long as the indi-
vidual continues to work (Friedberg and Webb 2005). Further, although
this would not have to be the case, de ned contribution plans tend to
be less generous than de ned bene t plans in practice. Ghilarducci and
Sun (2006) nd that employers contribute signi cantly less per capita
under de ned contribution plans than under de ned bene t plans. Simi-
larly, the cash balance plans offered by employers who have converted
from traditional de ned bene t plans generally have been less gener-
ous than the plans they replaced (Government Accountability Of ce
2005a). Research has concluded that participants in de ned contribu-
tion plans retire later on average—perhaps as much as two years lat-
er—than participants in de ned bene t plans and that the shift towards
de ned contribution plans has been an important reason for the recent
increases in labor force participation at older ages (Friedberg and Webb
2005; Munnell, Cahill, and Jivan 2003).
New rules regarding phased retirement plans included in the Pension
Protection Act of 2006 also may increase the share of older individuals
who choose to work.
The new rules allow in-service distributions from
tb08fogjch5.indd 169tb08fogjch5.indd 169 9/10/2008 12:47:10 PM9/10/2008 12:47:10 PM
170 Abraham and Houseman
de ned bene t pension plans to employees aged 62 and older, meaning
that they are allowed to reduce their hours on the job while beginning
to collect pension bene ts, something that had not previously been per-
mitted for employees below the normal retirement age speci ed in their
employers pension plan.
Falling Retiree Health Bene t Coverage
Paralleling these changes in pension bene ts has been a decline in
the coverage and generosity of retiree health insurance. As medical-
care costs have risen, so too have the costs of retiree health insurance.
In addition, the rules that govern how rms must account for those costs
have changed in a way that makes it less attractive for rms to offer
a retiree health insurance plan. Federal Accounting Standards Board
(FASB) Statement of Financial Accounting Standards (SFAS) No. 106,
issued in 1990, requires companies to report retiree health insurance
bene ts as a liability on their nancial accounting statements.
Issuance of this standard is widely believed to have triggered a
reduction in the coverage of such plans. Data from employer surveys
sponsored by the Kaiser Family Foundation show that the fraction of
companies with 200 or more employees offering retiree bene ts fell
from 66 percent in 1988 to 36 percent in 1993 and has uctuated be-
tween 35 and 40 percent since that time, except for a dip in 2005 (John-
son 2007; McCormack et. al 2002). Other data from the Medical Ex-
penditure Panel Survey Insurance Component (MEPSIC), an employer
survey that covers both large and small employers, show that the share
of private sector employees who work for an employer with a retiree
health plan fell between 1997 and 2003 (Buchmueller, Johnson, and
Lo Sasso 2006). Furthermore, in recent years a growing share of rms
with plans have closed them to new retirees (Eibner, Zawacki, and Zim-
merman 2007). In coming years, all of these changes can be expected
to reduce signi cantly the share of retirees with employer-sponsored
health coverage. In addition, the premium contributions that eligible
workers must pay to be covered have risen sharply (Buchmueller, John-
son, and Lo Sasso 2006). Increases in the cost of coverage may strain
the nances of older individuals and also may reduce take-up rates,
further depressing the share of retirees with employer-provided health
insurance coverage.
tb08fogjch5.indd 170tb08fogjch5.indd 170 9/10/2008 12:47:10 PM9/10/2008 12:47:10 PM
Removing Barriers to Work for Older Americans 171
Inadequate Savings Rates
The ongoing shift away from de ned bene t plans and anticipated
declines in retiree health insurance coverage imply that fewer Ameri-
cans will enjoy generous company-provided bene ts in the future, and
hence that they will need to save more to ensure adequate income in
retirement. Yet a recent survey of adults conducted by the Employee
Bene t Research Institute (EBRI) shows that many do not fully under-
stand how these changes affect their retirement security, and, among
those who do, few have altered their savings behavior to compensate
for the decline in generosity of retirement income provided by employ-
ers. EBRI found that large numbers of adults grossly underestimated the
amount of savings they would need to cover expected medical expenses
in old age and overestimated their chances of receiving traditional de-
ned bene t pension plans (Helman, Copeland, and VanDerhei 2006).
More generally, recent studies have highlighted the fact that many
individuals are not saving suf ciently to maintain living standards in
old age (Congressional Budget Of ce 2003). One effective method of
increasing retirement savings would be to change default options to
require employees who do not wish to participate in 401(k) and similar
retirement plans to opt out rather to require those who wish to par-
ticipate to opt in (Madrian and Shea 2001). The Pension Protection Act
creates a safe harbor for such automatic enrollment plans in the form of
minimum employer contribution schedules that, if adopted, exempt the
employer from the usual nondiscrimination tests otherwise required to
ensure that 401(k) plans do not offer disproportionate bene ts to high-
ly paid employees (Deloitte 2006). This safe harbor provision should
encourage more employers to adopt an automatic enrollment default.
Another proposed reform would replace incentives for retirement sav-
ings that operate by making contributions to retirement plans tax de-
ductible—an incentive that is worth considerably more to high-income
households than to low- or middle-income households because of their
higher marginal tax rates—with a plan that matches contributions up to
some threshold amount (Gale, Gruber, and Orszag 2006).
It is unclear how effective such reforms can be in increasing retire-
ment savings for the typical American. Their adoption would not change
the fundamental fact that wages for middle- and lower-income Ameri-
cans have been stagnant or falling, and that many feel they have limited
tb08fogjch5.indd 171tb08fogjch5.indd 171 9/10/2008 12:47:10 PM9/10/2008 12:47:10 PM
172 Abraham and Houseman
slack in their household budgets to set aside money for retirement. This
problem is implicit in the analysis of Scholz, Seshadri, and Khitatrakun
(2006), who challenge the conventional wisdom that Americans are not
saving optimally for retirement. For individuals with low or modest
incomes, “optimal” savings may be at or near zero with standard dis-
count rates. Instead of relying on savings, many Americans, because of
declining Social Security and pension bene ts, may need to delay full
retirement to maintain their living standards.
The premise that there will be a growing supply of older Americans
who want to work seems uncontroversial. Recent Social Security and
pension policy reforms, coupled with inadequate retirement and sav-
ings income and the decline in retiree health bene ts, will give more
older Americans an incentive to postpone retirement or reenter the labor
force. By itself, an increase in the number of older workers seeking em-
ployment will put downward pressure on their wages or, if wages do not
adjust smoothly, will result in involuntary or disguised unemployment.
Employer Demand for Older Workers
Some have argued that any increase in the supply of older Ameri-
cans seeking employment will be matched or exceeded by an increase
in demand among employers seeking to hire them. Many policy ana-
lysts predict that, faced with massive retirements among the baby boom
generation, employers soon will face serious labor shortages. This,
they believe, will induce employers to work harder to retain their ex-
isting workforce and to recruit more actively among the pool of retir-
ees (AARP 2006; Dychtwald, Erickson, and Morison 2006; Ernst and
Young 2006; Judy and D’Amico 1997). These analysts predict that em-
ployers’ need for older workers will lead them to be more accommo-
dating both of older workers’ physical limitations and of their desires
for more exible employment, including part-time or part-year work.
Under this scenario, older workers will nd it relatively easy to arrange
tb08fogjch5.indd 172tb08fogjch5.indd 172 9/10/2008 12:47:10 PM9/10/2008 12:47:10 PM
Removing Barriers to Work for Older Americans 173
exible, phased retirement with their existing employer or to land an
attractive job with a new employer.
Other researchers are more skeptical that the future will be this rosy
for seniors. Although the baby boom generation already has started to
retire, existing research nds little evidence that companies are moving
to establish broad-based worker retention programs (AARP 2006; Ernst
and Young 2006; Government Accountability Of ce 2005b). A large
employer survey about phased retirement options offered to white-col-
lar workers found that such programs were uncommon and that, when
these programs were offered, employers typically operated informally,
choosing to make special arrangements for the most valued employees
(Hutchens 2007). Dychtwald, Erickson, and Morison (2006), who make
a strong case that employers will need to retain employees to avoid
shortfalls in the near future, also acknowledge that employers will want
to be selective in whom they retain. For this reason, they advocate liber-
alization of “nondiscrimination tests for exible retirement plans so that
employers [can] more easily customize work arrangements and offer
them to employees with exceptionally valuable skills and experience
without breaking antidiscrimination laws and uniformity mandates” (p.
62). Low-skilled workers, who are the most vulnerable to cuts in Social
Security and in the worst position to increase their own savings for re-
tirement, arguably are the least likely to bene t from any growth in em-
ployer retention programs. Freeman (2006) also has criticized the labor
shortage hypothesis, arguing that employers will meet any shortages
with immigration and offshoring. Again, the least skilled workers are of
greatest concern, because they tend to be the most adversely affected by
pressures from immigration and offshoring.
Evidence on the Dif culty Older Workers Face in Finding
New Work
Many older Americans work long hours or in physically demand-
ing jobs and would like to cut back on their hours or change the type of
work they do as they age. Although much of the literature on retirement
transitions discusses the prevalence of so-called bridge jobs, it appears
that far more individuals would like to work in later years than in fact
do so. For example, 73 percent of workers aged 51 to 61 surveyed in the
1992 Health and Retirement Study (HRS) said that they would like to
tb08fogjch5.indd 173tb08fogjch5.indd 173 9/10/2008 12:47:10 PM9/10/2008 12:47:10 PM
174 Abraham and Houseman
continue paid work following retirement (AARP 1998). Other surveys
have yielded similar ndings. Yet actual employment rates among older
Americans are far lower than one might expect from these survey re-
sponses. In 2001, when those interviewed for the 1992 HRS would have
been aged 60 to 70, only 57 percent of men and 44 percent of women
aged 60 to 64, and only 32 percent of men and 21 percent of women
aged 65 to 69 were employed either part-time or full-time. Other data
show that among men aged 55 to 64 who received pension or retire-
ment-plan income in 2002 just over a third were working in March
2003, and the corresponding share among men 65 and older was only
12 percent (Purcell 2005).
Although there are multiple reasons why the number of individuals
who plan to work in later years is substantially higher than the number
who, in fact, realize their plans, our research suggests that the dif culty
older workers experience with transitioning to new jobs is a signi cant
factor. We draw on panel data from the Health and Retirement Study.
The HRS panel includes a representative sample of Americans born
in the years 1931 to 1941. Seven waves with interviews conducted
biennially from 1992 to 2004 are currently available. In each wave,
survey participants answer detailed questions about many aspects of
their health, work, and nances. We use questions about future plans
for work and retirement that were asked of individuals employed at the
time of the survey. We then exploit the panel structure of the data to
examine whether individuals followed through on these plans.
Among those with de nite future work or retirement plans, a minor-
ity (39 percent) indicated that they planned to stop work altogether, and
almost as many indicated that they planned to work fewer hours (29
percent) or change the type of work they do (7 percent) (Table 5.1). Ex-
ploiting the fact that the HRS reinterviews the same individuals at two-
year intervals, we examine the degree to which individuals followed
through on these work and retirement plans. Speci cally, we identify
the subset of individuals who indicated during the interview that they
planned to stop work altogether, cut back on their hours, or change the
type of work they do within the following two years. We then look at
whether what they are doing at the subsequent two interviews—which
take place about two years and about four years later—is consistent
with their stated plans. We classify people as having reduced their
weekly hours if the sum of weekly hours worked on all jobs dropped by
tb08fogjch5.indd 174tb08fogjch5.indd 174 9/10/2008 12:47:11 PM9/10/2008 12:47:11 PM
Removing Barriers to Work for Older Americans 175
eight hours or more between waves. Whether individuals have changed
the type of work they do is somewhat subjective, and there is no clean
measure of such a work change in the HRS. We consider anyone who
changed occupation to have changed the type of work they were doing.
Because our measure of work and retirement plans groups those who
plan to change their type of work with those who plan to begin working
for themselves, we also treat those who move from employee to self-
employed status, or the reverse, as having changed their type of work.
Table 5.2 compares individuals’ work and retirement plans with
what they were actually doing when next interviewed two years later.
Most interesting is the fact that those with near-term plans to stop work
altogether were much more likely to follow through on those plans
than were individuals who planned to reduce their hours of work or
change the type of work they do. Whereas about 65 percent of those
who planned to retire fully were not working at the next interview
about two years later, only 35 percent of those who planned to reduce
their hours and 24 percent of those who planned to change the type of
work they do followed through on their plans. Roughly half of those
who failed to realize plans to reduce their hours or change their type
of work made no changes in their employment situation and roughly
half stopped working altogether. Although various factors, such as an
unanticipated worsening of health status, may help explain why rela-
tively few follow through on these stated plans, analysis we have done
suggests that, for many older workers, the need to change jobs is an
Table 5.1 Plans for Work and Retirement
Percent of observations for which plan reported
“Don’t know”
“Don’t know”
Stop work altogether 24.5 39.3
Work fewer hours 18.6 29.8
Change type of work 4.5 7.3
Never stop work 7.8 12.4
Other 7.0 11.2
Don’t know 37.7 n.a.
SOURCE: Authors’ calculations from waves 1–7 of the Health and Retirement Study
(1992–2004), based on 18,758 reports of individuals’ work and retirement plans. All
estimates constructed using HRS sample weights.
tb08fogjch5.indd 175tb08fogjch5.indd 175 9/10/2008 12:47:11 PM9/10/2008 12:47:11 PM
Table 5.2 Labor Market Outcomes among Those with Near-Term Plans to Stop Work, Work Fewer Hours, or
Change Type of Work (%)
Work or retirement outcome after 2 years
Change planned
within 2 years Realized plan
hours of work
type of work
Reduced hours
and changed
type of work Stopped work No changes N
Stop work altogether 64.8 9.7 2.5 8.6 64.8 14.5 751
Work fewer hours 34.9 25.5 6.9 9.4 27.7 30.5 569
Change kind of work 24.4 12.3 6.5 17.9 33.9 29.5 123
SOURCE: Authors’ calculations from waves 1–7 of the Health and Retirement Study. Calculations based on reports from individuals
who indicated that they planned to stop work or make signi cant changes to hours worked or job held within two years and who were
interviewed in the following wave. Information on plans collected in waves 1 through 6; information on outcomes collected in following
wave. All estimates constructed using HRS sample weights. In computing the percent distribution of work and retirement outcomes for
people with given plans, we excluded a small number of cases (2 percent or fewer) for which the outcome could not be determined.
tb08fogjch5.indd 176tb08fogjch5.indd 176 9/10/2008 12:47:11 PM9/10/2008 12:47:11 PM
Removing Barriers to Work for Older Americans 177
important impediment to reducing hours and changing the type of work
they do. In Abraham and Houseman (2005) we show that, among the
employed, those who reported that their employer allowed employees
to reduce work hours were much more likely to plan to reduce work
hours and, if they had such plans, were much more likely to realize their
plans to reduce hours than were individuals whose employers did not
allow such exibility.
Because it may take more time than initially anticipated to make a
planned change, especially when the change entails nding a new job,
we also examine whether individuals realized their stated plans either
by the next wave interview, about two years later, or by the subsequent
interview, about four years later.
When examined over this longer pe-
riod, a higher percentage of individuals realized plans to reduce their
work hours or change their type of work. Even over this longer hori-
zon, however, only half of those planning to reduce their hours and a
third of those planning to change the type of work they do appear to
have followed through on their plans, compared to about 80 percent
who followed through on plans to stop work altogether. Four years after
stating plans to reduce their hours or change their type of work, those
who failed to realize these plans were roughly twice as likely to have
stopped working altogether as they were to be still working.
In analysis not reported here, we nd that adults with low education
levels have the most dif culty making such job transitions, as is consis-
tent with prior research on dislocated workers discussed below. Less-
educated individuals were signi cantly less likely to formulate plans
to reduce their hours of work or change the type of work they do, and,
conditional on having such plans, were signi cantly less likely than
more-educated workers to follow through on those plans. We estimate
that those who have not completed high school were 11 to 17 percent-
age points less likely to follow through on plans to reduce their hours
and 10 to 17 percentage points less likely to follow through on plans to
change the type of work they do.
Declining Job Stability and Its Consequences
The apparent decline in job stability in recent years poses a further
obstacle for older workers who wish to continue working. Although
long-term jobs are still quite common among older workers (Stevens
tb08fogjch5.indd 177tb08fogjch5.indd 177 9/10/2008 12:47:11 PM9/10/2008 12:47:11 PM
178 Abraham and Houseman
2006), recent evidence suggests that they are less common than they
used to be. In a detailed look at trends in employment over the last 50
years, Farber (2006) nds that mean and median job tenure among men
have declined for all age groups, as has the incidence of being in a long-
lasting job (one that lasts at least 10 or 20 years). The largest declines
have been among older men. Comparing men born in the 1930s to those
born in the 1950s, for example, Farber shows that median job tenure at
age 50 decreased by more than two years, from 11.9 years to 9.7 years.
In addition, Farber nds some evidence of job churning (which he de-
nes as jobs lasting less than a year) among men in their thirties. Men
in their twenties always have experienced a lot of job turnover, a fact in-
terpreted to mean that they were trying out various jobs before settling
in to long-term employment. The fact that job churning has increased
signi cantly among men in their thirties suggests that men today may
be having more dif culty nding a suitable long-term job than were
men in previous generations.
Research evidence suggests that at least some of the decline in job
stability among older workers is associated with layoffs. While the inci-
dence of layoff is lower among older than among younger workers, the
gap has narrowed in recent recessions (Farber 2005), and, conditional
on tenure in the job, workers aged 45 to 64 have become substantially
more susceptible to job loss (Farber 2007).
The narrowing gap between the dislocation rates of younger and
older workers may be related to an aging workforce in sectors—like
manufacturing—that are especially sensitive to business cycles and that
have experienced large secular declines in employment in recent years.
Case study evidence also suggests that business perceptions about the
desirability of retaining older workers when there is a layoff may be
changing. Historically, layoffs generally occurred in inverse seniority
order, even in the absence of union contracts dictating such layoff rules
(Abraham and Medoff 1984). This pattern has been widely interpreted
to imply that employers value the skills and experience of their long-
tenure employees, though other research has indicated that the higher
pay of more senior workers cannot be fully justi ed on the basis of
their higher productivity (Medoff and Abraham 1980, 1981; Hellerstein
and Neumark 2007). Recent high-pro le cases involving Wal-Mart and
Circuit City indicate that at least some employers have concluded that
the extra pay long-tenure employees typically receive exceeds any ad-
tb08fogjch5.indd 178tb08fogjch5.indd 178 9/10/2008 12:47:11 PM9/10/2008 12:47:11 PM
Removing Barriers to Work for Older Americans 179
ditional productivity garnered from their experience and have adopted
policies that some allege discourage long-term employment (Green-
house and Barbaro 2006; Leonhardt 2007). Besides wages that arguably
rise faster than productivity, the high cost of providing health insurance
for older workers provides an incentive to companies to lower the aver-
age age at the workplace.
Taken as a whole, the evidence of a decline in job stability implies
that older workers will be less able to rely on long-term jobs that last
until retirement and more likely to need to search for new jobs, possibly
in new elds, late in their working life. Yet research on dislocated work-
ers—de ned as those who have lost their job for economic reasons—
unambiguously shows that older workers have an especially dif cult
time making job transitions. Using data from the Health and Retirement
Study, Chan and Stevens (2001) nd large and long-lasting adverse ef-
fects of job loss on employment among older workers. Only 61 percent
of men and 55 percent of women who involuntarily lose their jobs are
reemployed two years following job loss. Compared to similar workers
who do not lose their jobs, individuals who experience job loss at age
55 are an estimated 20 percentage points less likely to hold a job at age
59. Farber (2005) nds similarly low levels of reemployment among
older, displaced workers. He also nds much lower reemployment rates
among less-educated workers. Among displaced workers who become
reemployed, older displaced workers, who on average had longer ten-
ure and were earning more on the job they lost, also experience substan-
tially higher earnings losses.
What happens to older dislocated workers who fail to nd new
jobs? Many appear to wind up in the Social Security Disability Insur-
ance (SSDI) program, collecting bene ts from SSDI until they qualify
for regular, age-related Social Security bene ts. The fraction of the
U.S. adult population on SSDI has grown dramatically over the last 20
years, increasing from 2.2 percent of 25-to-64-year-olds in 1985 to 4.1
percent in 2005. The largest increases have occurred among older, less-
educated subgroups. Using data from the Survey of Income and Program
Participation (SIPP), Autor and Duggan (2006) estimate that between
1984 and 2004 the rate of SSDI receipt rose from 14.8 to 19.7 percent
among male high school dropouts aged 55 to 64 and from 9.1 to 12.7
percent among female high school dropouts of the same ages. The ap-
proximately 5-percentage-point increase in disability coverage among
tb08fogjch5.indd 179tb08fogjch5.indd 179 9/10/2008 12:47:12 PM9/10/2008 12:47:12 PM
180 Abraham and Houseman
high school dropout men aged 55 to 64 compares to an 8-percentage-
point decline in the labor force participation rate in this age group over
the same period. Autor and Duggan point to increases in the generosity
of bene ts and to changes that have made it easier to qualify for bene-
ts when an individual has dif culty nding employment—not true in-
creases in the incidence of disability—as explanations for the dramatic
growth in the program. They argue cogently that SSDI has become a
program for the unemployable.
The above evidence indicates that older workers, particularly less
skilled and less educated workers, have dif culty transitioning to new
jobs. Part of the problem undoubtedly is that many of these workers do
not know how to go about looking for a job. And even before they start,
many are convinced that employers simply are not interested in hiring
older workers, which itself may discourage active search (Government
Accountability Of ce 2005b).
An audit study conducted by Lahey (2008) provides perhaps the
best research evidence to date of employer discrimination against older
workers applying for entry-level jobs. In her study, Lahey sent resumes
of women to be considered for entry-level positions to prospective
employers in Boston, Massachusetts, and St. Petersburg, Florida. The
resumes were carefully written to appear nearly identical, except that
information on the resumes indicated that some applicants were rela-
tively young (age 35 or 45), while others were older (age 50, 55, or 62).
Younger applicants were more than 40 percent more likely than older
applicants to be called in to interview for the position. Two commonly
offered reasons why employers, all else being the same, might prefer
younger to older workers are that older workers lack necessary skills,
especially technical skills, and that their wage and bene ts costs are too
tb08fogjch5.indd 180tb08fogjch5.indd 180 9/10/2008 12:47:12 PM9/10/2008 12:47:12 PM
Removing Barriers to Work for Older Americans 181
Lack Necessary Skills
Although skills and experience are much-touted assets of older
workers (see, for instance, AARP 2005), years of work experience may
bene t an older worker primarily in the job in which he or she accrued
that experience. Unless the skills from one job are readily transferable
to another job, years of accrued experience provide an older worker
with little advantage when those workers want or need to change jobs.
Moreover, businesses indicate one problem they have with hiring old-
er workers is that they often lack up-to-date technical skills (Arizona
Mature Worker Initiative 2006). Older workers themselves sometimes
acknowledge their lack of computer skills (Government Accountability
Of ce 2005b). Employers may be reluctant to invest in training older
workers given the relatively short time that they can be expected to con-
tinue working (Arizona Mature Worker Initiative 2006). Perceptions
that older workers are more rigid and slower to learn may reinforce
employers’ unwillingness to hire older workers who require training
(Bendick, Jackson, and Romero 1996).
Beyond lacking technical skills, older job applicants may not pos-
sess physical or social attributes that are important in some jobs. Older
workers may no longer have the physical stamina or dexterity to han-
dle certain jobs, particularly jobs in low-skilled manual occupations in
which many less-educated individuals historically have been employed.
In addition, employers believe that some older workers would nd it
dif cult to accept being part of an ethnically or culturally diverse work-
force, meaning that their presence could undermine effective commu-
nication and smooth functioning in many workplace settings (Arizona
Mature Worker Initiative 2006).
High Wage and Bene ts Costs
Based on analyses of personnel records for the salaried workforces
of several large companies, Medoff and Abraham (1980, 1981) nd that
senior workers receive higher pay than equally productive junior work-
ers. Hellerstein and Neumark (2007) nd that manufacturing workers
aged 55 and over are less productive than younger workers, but this
lower productivity is not matched by lower earnings. The high earn-
ings of older workers relative to their productivity is widely seen as an
tb08fogjch5.indd 181tb08fogjch5.indd 181 9/10/2008 12:47:12 PM9/10/2008 12:47:12 PM
182 Abraham and Houseman
impediment to the hiring of older workers in many developed countries
(OECD 2006). Recent high-pro le cases in the United States suggest
that relatively high pay for workers with long job tenures is becoming
an important human resources issue for cost-conscious corporations. A
memorandum by a top Wal-Mart executive to its board of directors, for
instance, outlined a proposed policy to increase turnover and thereby
reduce the average tenure of its stores’ employees, which it calls asso-
ciates, by setting wage caps on certain positions and requiring staff to
work nights and weekends. The memo noted that the cost of an associ-
ate with seven years of tenure was 55 percent more than the cost of an
associate with one year of tenure, yet the productivity of the two was
the same. Similar concerns at Circuit City led to dismissals of about 8
percent of its employees who were deemed overpaid so that these em-
ployees could be replaced by lower-cost workers (Leonhardt 2007).
Although human resource practices that reward employee perfor-
mance with pay raises that exceed employees’ productivity growth may
be coming to an end, the apparent pervasiveness of the practice points
to another reason why employers are reluctant to hire older workers as
new employees. According to one survey, employers believe that many
older workers expect higher salaries than those that come with the jobs
for which they are applying (Arizona Mature Worker Initiative 2006).
A corollary is that employers fear that, if older workers take a job with
lower wages than they feel entitled to, they may view the relatively low
wages as unfair; consequently they may have low morale and perform
poorly on the job. Thus, employers may be reluctant to hire a person
who has earned signi cantly higher wages in the past, even if that indi-
vidual indicates a willingness to accept the job.
Older workers also are more expensive to insure. Health insurance
has become the most costly worker bene t (Government Accountabil-
ity Of ce 2006a), and according to analysis by Towers Perrin for the
AARP (AARP 2005), on average, workers aged 50 to 65 have medical
expenses that are 1.4 to 2.2 times higher than workers in their thirties
and forties, translating into signi cantly higher health insurance costs.
One of the leading obstacles to hiring older workers, according to em-
ployer surveys, is the high cost of providing them with health insurance
(Arizona Mature Worker Initiative 2006). One study found that em-
ployers with health bene t plans are signi cantly less likely to hire per-
sons aged 55–64 than are employers who do not offer health insurance
tb08fogjch5.indd 182tb08fogjch5.indd 182 9/10/2008 12:47:12 PM9/10/2008 12:47:12 PM
Removing Barriers to Work for Older Americans 183
(Scott, Berger, and Garen 1995). And the issue of high health insurance
costs generally does not go away once the older worker reaches age
65 and thus quali es for Medicare coverage. Under existing Medicare
rules, if an employee aged 65 or over working for an employer with 20
or more employees is covered by employer-provided health insurance,
in most cases that insurance policy, not Medicare, is the rst payer for
any claim that may be made.
The aging of the American population is expected to place severe
strains on the Social Security and Medicare systems in the coming de-
cades and has led to a broad consensus that it is in the public interest
to increase employment among older Americans. The perceived need
to increase employment among older Americans comes at a time when
their job tenure is on the decline and job loss is on the rise. Although
labor force participation among older women has risen in recent de-
cades and labor force participation among men over age 65 has edged
up since the mid-1990s, labor force participation among men in their
fties—i.e., men in their immediate preretirement years—has declined
over the last two decades. The decline has been particularly marked for
less-educated men.
As discussed earlier in the chapter, selected policies designed to pro-
vide greater incentives for older Americans to work have been adopted.
Most notably, the age at which individuals will qualify to retire with full
Social Security bene ts is rising from age 65 to age 67. For those with
traditional de ned bene t pensions, reforms to ERISA will facilitate
workers’ participation in phased retirement programs within their com-
pany. Still, for many workers, whether because they seek a change in
hours or job duties or because they are forced out of their jobs, contin-
ued employment at older ages will require a change of employers. The
failure by many to make these job transitions has signi cant, if hidden,
public costs, which include growing dependence of older Americans on
the Social Security system and other forms of public assistance. Poli-
cies that provide nancial incentives to older Americans to work, we
believe, must be supplemented by policies that ease job transitions and
tb08fogjch5.indd 183tb08fogjch5.indd 183 9/10/2008 12:47:13 PM9/10/2008 12:47:13 PM
184 Abraham and Houseman
remove some of the substantial barriers to being hired that older Ameri-
cans face. Below we review federal employment and training programs
that serve older workers and recent initiatives that some states have
taken to facilitate employment among older Americans. We conclude
by outlining policies that we believe are needed to ll existing gaps.
Federal Programs
The Senior Community Services Employment Program (SCSEP)
is the only federal employment program that speci cally targets old-
er Americans. SCSEP serves approximately two-thirds of individuals
aged 55 and over who receive federal employment and training services
(General Accounting Of ce 2003). Funded under the Older Americans
Act of 1965, the SCSEP program provides subsidized part-time and
community service employment for individuals aged 55 and older who
have incomes below 125 percent of the poverty level. Historically,
moving participants to unsubsidized employment was not a stated goal
of the program. Although this was changed in 2000, placement rates
into unsubsidized employment remain relatively low (Government Ac-
countability Of ce 2006b). According to of cial program statistics, in
recent years SCSEP has served around 100,000 seniors nationwide (Ta-
ble 5.3), most of them over age 65. The U.S. Department of Labor esti-
mates that this represents fewer than 1 percent of the eligible population
(Government Accountability Of ce 2006b). Program funding has been
stagnant in current dollars and, as shown in Figure 5.3, has been falling
in real terms since the late 1990s.
The Work Force Investment Act (WIA) is the country’s primary em-
ployment and training services program, although the number of older
workers who are served by WIA programs is considerably smaller than
the number served by SCSEP (Table 5.3).
WIA provides basic labor
market information and preliminary job skills services to any adult 18
or over who seeks them. More intensive job search assistance and train-
ing services are limited to individuals enrolled in one of three funded
WIA programs: the adult, dislocated worker, and youth programs. Older
Americans are served under the rst two programs. Priority for enroll-
ment in the WIA adult program is given to individuals on public as-
sistance and others in low-income households. Enrollment in the WIA
tb08fogjch5.indd 184tb08fogjch5.indd 184 9/10/2008 12:47:13 PM9/10/2008 12:47:13 PM
Removing Barriers to Work for Older Americans 185
dislocated worker program is reserved for individuals permanently laid
off from jobs.
In July 2000, the WIA adult and dislocated worker programs re-
placed similar programs funded under the Job Training Partnership
Act (JTPA). The JTPA dislocated worker program enjoyed funding in-
creases that offset declines in funding for the JTPA adult program over
much of the 1990s, but under WIA, funding for both programs has been
stagnant in current dollars and falling in real dollars. In constant dollars,
total funding for the WIA adult and dislocated worker programs was 35
percent lower in 2007 than funding for the comparable JTPA programs
had been ten years earlier, and the decline has been even steeper relative
to the size of the working-age population. Within this funding context,
of cial program statistics show that the absolute number of workers
over age 55 served by WIA programs has been lower than the number
served under the corresponding JTPA programs and that older workers
have declined as a fraction of all adults served.
Figure 5.3 Trends in Real Spending on Employment and Training
Programs (2006 constant dollars)
SOURCE: Budget numbers from U.S. Department of Labor, Employment and Train-
ing Administration, Training and Employment Programs Budget Authority. PY 1984–
2004, The authors de ated all budget g-
ures using the CPI Series CUUROOOOSAO.
tb08fogjch5.indd 185tb08fogjch5.indd 185 9/10/2008 12:47:13 PM9/10/2008 12:47:13 PM
Table 5.3 Data on Program Exiters, JTPA, WIA, and SCSEP Programs, 1997–2004
Program year
1997 1998 1999 2000 2001 2002 2003 2004
SCSEP program
Program exiters
96,852 105,829 93,000
As % of population 55–64 0.38 0.38 0.32
JTPA/WIA adult programs
Program exiters
159,389 163,223 142,147 109,868 193,518 261,544 229,607 225,683
As % of population 0.10 0.10 0.09 0.07 0.11 0.15 0.13 0.13
Program exiters age 55+
14,823 15,180 14,073 6,922 11,418 14,385 13,776 13,541
As % of population 55–64 0.07 0.07 0.06 0.03 0.05 0.06 0.05 0.05
As % of all exiters 9.30 9.30 9.90 6.30 5.90 5.50 6.00 6.00
JTPA/WIA dislocated worker programs
Program exiters
266,112 240,896 205,637 99,491 147,715 200,264 194,425 178,446
As % of population 0.16 0.15 0.12 0.06 0.09 0.12 0.11 0.10
Program exiters age 55+
26,611 24,812 22,826 11,441 16,692 22,229 21,970 21,592
As % of population 55–64 0.12 0.11 0.10 0.05 0.07 0.09 0.08 0.07
As % of all exiters 10.00 10.30 11.10 11.50 11.30 11.10 11.30 12.10
tb08fogjch5.indd 186tb08fogjch5.indd 186 9/10/2008 12:47:13 PM9/10/2008 12:47:13 PM
SOURCE: Data on programs from Social Policy Research Associates (PY 1997–2000), the 2001 WIASRD Data Book, http://www.doleta
.gov/performance/results/PY_2001_WIASRD_Databook.pdf (PY 2001), and the corresponding data books and links for PY2002 through
JTPA/WIA adult plus dislocated worker
Program exiters
425,501 404,119 347,784 209,359 341,233 461,808 424,032 404,129
As % of population 0.26 0.24 0.21 0.12 0.20 0.27 0.24 0.22
Program exiters age 55+
41,434 39,992 36,898 18,363 28,109 36,614 35,746 35,133
As % of population 55–64 0.19 0.18 0.16 0.08 0.12 0.14 0.13 0.12
As % of all exiters 9.74 9.90 10.61 8.77 8.24 7.93 8.43 8.69
Trade adjustment assistance programs
Program exiters
26,363 24,883 30,047
As % of population 0.01 0.01 0.02
Program exiters age 55+
3,045 2,924 3,933
As % of population 55–64 0.01 0.01 0.01
As % of all exiters 11.55 11.75 13.09
tb08fogjch5.indd 187tb08fogjch5.indd 187 9/10/2008 12:47:14 PM9/10/2008 12:47:14 PM
188 Abraham and Houseman
Under JTPA, states were required to set aside 5 percent of their
adult program allotment for older workers. This provision was dropped
in WIA and, as can be seen in Table 5.3, the share of those in the adult
program who are aged 55 and over declined from about 9 percent in
the JTPA adult program to about 6 percent in the WIA adult program.
Although one might hypothesize that eliminating the quota on spend-
ing for older workers in WIA resulted in a more ef cient allocation
of resources, recent government reports have raised concerns that the
federal performance standards used to evaluate organizations that ad-
minister the WIA programs have resulted in a bias against serving older
workers. Until 2005, the performance standards for WIA service pro-
viders included measures of participants’ postprogram earnings relative
to their preprogram earnings. Some observers think that this created
a disincentive to serve older workers, many of whom have consider-
able prior work experience and may wish to transition from full-time to
part-time employment (General Accounting Of ce 2003). In an attempt
to address this problem, revised performance standards introduced in
2006 substitute a measure of postprogram earnings for the previous
earnings change measure (Employment and Training Administration
2006). However, even this new measure may discourage providers from
serving older workers, who are likely to seek part-time work and thus
have relatively low earnings.
Older workers also may receive job search assistance and training
under Trade Adjustment Assistance programs, which are reserved spe-
ci cally for workers displaced by foreign trade. Fewer than 4,000 TAA-
quali ed workers—about 13 percent of all participants in TAA training
and placement programs—were 55 or older in 2003. The Trade Act of
2002 wrote into law a ve-year demonstration program under which
eligible workers aged 55 and older who agree to forgo TAA-funded
training and are able to nd jobs within 26 weeks that pay less than their
previous earnings (and less than $50,000) are eligible for a wage sub-
sidy of up to $10,000 cumulatively to supplement their earnings. States
moved slowly to implement the demonstration program and, as was
the case with a similar Canadian program in place from 1995 to 1998
(Levine 2007), take-up of the TAA wage subsidies has been relatively
low. In addition to a lack of awareness of the program and not wishing
to forgo the possibility of training, eligible workers who did not partici-
tb08fogjch5.indd 188tb08fogjch5.indd 188 9/10/2008 12:47:14 PM9/10/2008 12:47:14 PM
Removing Barriers to Work for Older Americans 189
pate cite the dif culty of nding a job within six months as required by
program rules (Government Accountability Of ce 2006c).
Recent State Initiatives
Although the federal government has taken no major policy initia-
tive for older workers in recent years, a few states are beginning to take
steps to address what is perceived as a growing need for employment
services among this population. To identify innovative policies at the
state level, we conducted an informal survey of relevant state workforce
agencies through their national association (the National Association
of State Workforce Agencies). Although most states do not have older
worker programs besides the federally funded SCSEP, 10 states re-
sponded to our survey with information on initiatives they are currently
taking or planning. In some instances, the AARP has worked with states
to plan and implement initiatives. In addition, the National Governors
Association has selected eight states to participate in a policy academy
designed to develop model programs to meet the needs of mature work-
ers (National Governors Association 2007).
These state initiatives generally fall into three broad categories. The
rst is employer outreach and education. Several states have initiated
programs to advise employers on how to accommodate an aging work-
force. Initiatives also include information campaigns to combat what
are regarded as inaccurate and damaging stereotypes of older workers
with the aim of reducing discriminatory practices against those work-
ers. These public relations efforts emphasize positive attributes of older
workers relative to young workers, such as reliability and good social
skills. Patterned in part on an AARP program, efforts in several states
also feature or are planning to feature employers who evidence a com-
mitment to hiring and promoting older workers. The idea behind these
programs is at one level to help seniors connect with employers who
are willing to hire them, but at a deeper level to provide positive public
relations for the identi ed companies and advance the notion that hiring
older workers is a good business practice.
A second area involves outreach and better delivery of existing ser-
vices to older workers. One of the most common initiatives adopted
by states is to place an older worker specialist in the one-stop centers,
which serve as the central clearinghouse for all workforce development
tb08fogjch5.indd 189tb08fogjch5.indd 189 9/10/2008 12:47:14 PM9/10/2008 12:47:14 PM
190 Abraham and Houseman
programs under WIA. This specialist would be able to better direct older
workers to services meeting their speci c needs and, some think, make
older workers feel more welcome at one-stops. The concern that older
workers are often reluctant to go to one-stop centers was expressed by a
number of state workforce representatives, some of whom consequent-
ly planned outreach efforts at senior centers and other places where
seniors might congregate. Such outreach may also involve providing
access to job listings at these remote sites.
A third area involves tailoring programs to meet speci c needs of
older workers. One program, for instance, emphasizes peer counsel-
ing, networking, life planning seminars, and other resources targeting
those over 50 who are changing jobs and even careers. In addition to
having special planning and counseling needs, older individuals often
lack basic technical and computer skills that are needed for many jobs,
and, indeed that may be needed even to nd a job. Training in Internet
job search and in the basic computer skills required on many jobs is
widely recognized as a prevalent need among older workers. Arizona,
for instance, is planning the development of a Web-based mechanism
for posting jobs and resumes that would be technologically friendly
for older workers and ef cient for businesses. Some states offer free or
reduced tuition to older adults who wish to advance their skills at public
postsecondary institutions.
In sum, states are experimenting with a variety of approaches to
increase workforce participation among seniors. These efforts involve
changing employer perceptions of older workers and hence receptive-
ness to hiring them, reaching out to older individuals who may be un-
aware of work opportunities, and tailoring programs to meet the special
needs of older workers. We are aware of no serious effort to evaluate
any of these initiatives, which, given the lack of federal funding or other
resources available, are all modest in scope.
Policy Recommendations
Several circumstances, discussed above, serve as a backdrop for
our policy recommendations: 1) there is a public interest in increasing
employment among older Americans, 2) many older workers will need
to make job changes late in life in order to remain employed, 3) older
workers face signi cant impediments to nding new employment, and
tb08fogjch5.indd 190tb08fogjch5.indd 190 9/10/2008 12:47:15 PM9/10/2008 12:47:15 PM
Removing Barriers to Work for Older Americans 191
4) public funding for employment and training programs that help older
workers has been low and is falling. These facts lead us to make the fol-
lowing recommendations:
Increase funding for employment and training programs to
help smooth employment transitions for older workers. While the
population of older workers is growing and many in this population
will need to make job transitions to remain employed, overall funding
for the major government programs designed to help workers prepare
for and nd employment has fallen by over a third in real terms in the
last decade, and older workers have received a shrinking share of that
shrinking pie. The chances that an older person who leaves or loses a
job will reenter employment are low unless the individual transitions
to a new job in a reasonably short period of time. Because many who
fail to nd jobs end up collecting Social Security disability payments
or some other form of public assistance, it is critical to have effective
programs to help older workers transition to new jobs.
To put the funding of workforce programs for older workers in some
perspective, in 2004 about 0.1 percent of the adult population aged 55
to 64 participated in a WIA adult or dislocated worker program.
the same year, about 9 percent of those aged 55 to 64 were collecting
SSDI, at an average annual cost per participant of two-and-a-half to
three times that of the average cost per participant of serving someone
in a WIA program. Although by no means does everyone who drops out
of the labor force qualify for SSDI, individuals who fail to nd employ-
ment are much more likely to receive other forms of public assistance,
such as food stamps, Medicaid, and public housing. Even a return-to-
work program with a modest success rate could save taxpayer dollars.
In the absence of better information on the return-on-investment
to be expected from increased expenditures on such programs, history
may provide some useful context. Returning real expenditures on the
employment and training programs that serve older workers (SCSEP,
the WIA adult program, the WIA dislocated worker program, and the
TAA program) to the level of a decade ago would require an increase
in total spending of about 40 percent. Taking into account the growth in
population that has occurred over the past 10 years and returning spend-
ing per working age adult (aged 25–64) to its 1997 level would require
an increase in spending of about 48 percent.
tb08fogjch5.indd 191tb08fogjch5.indd 191 9/10/2008 12:47:15 PM9/10/2008 12:47:15 PM
192 Abraham and Houseman
Modify WIA performance standards to eliminate disincentives
to serve older workers. The performance standards currently used to
evaluate the federally funded WIA programs include a measure of the
subsequent earnings of participants taken from state unemployment in-
surance quarterly wage records data. Because these data record only
total earnings in a quarter, not hourly wages, and because older work-
ers are more likely than younger workers to want part-time employ-
ment, this performance standard creates an unintended disincentive for
program operators to serve older workers. Although it is an improve-
ment upon the previously used measures that compared postprogram
earnings to preprogram earnings, the WIA earnings performance stan-
dard requires further modi cation to eliminate this disincentive. One
straightforward way to do this would be to develop different standards
to be applied to clients of different ages. For instance, performance
measures for employment services programs operating in the United
Kingdom and Australia explicitly take into account the fact that some
individuals face more serious barriers to employment than others; thus,
service providers receive more credit for placing clients with signi cant
employment barriers, such as older workers, into jobs (OECD 2006).
Tailor programs to meet needs of older workers. As a general
proposition, it is important that programs be designed appropriately to
meet the needs of special populations, including older workers. Experi-
mentation with initiatives for older workers such as those recently taken
or planned by some states should be encouraged. Promising approaches
include outreach and education efforts directed towards both employers
and seniors, placing older-worker specialists in one-stop employment
centers, and developing job-search assistance programs and job training
courses that address skill de ciencies common among seniors, such as
de ciencies in basic computer skills. Experiences in Australia suggest
that older people particularly bene t from intensive assistance that in-
cludes a personal coach and career counseling (OECD 2006).
Rigorously evaluate older-worker initiatives. While we are con-
dent that program providers can do a better job of serving older work-
ers and helping them to nd employment, it must be emphasized that
relatively little is known about how this can best be done. Rigorous
evaluation of new initiatives at the state level to assess their effective-
tb08fogjch5.indd 192tb08fogjch5.indd 192 9/10/2008 12:47:15 PM9/10/2008 12:47:15 PM
Removing Barriers to Work for Older Americans 193
ness and their suitability as models for the national level must be an in-
tegral part of any new funding for such programs. In the same spirit, the
planned evaluation of the wage subsidies for older workers, introduced
as part of the Trade Adjustment Assistance program, should be com-
pleted. Programs that prove to be cost-effective should be promoted
and expanded.
Reform health-care nancing to reduce disincentives to hiring
older workers. The high cost of providing health insurance for older
workers is a major impediment to reemployment in good, full-time jobs
for older workers. For those aged 65 and older, making Medicare rath-
er than the employer insurance policy the rst payer in the event of a
claim would help address this problem. However, such reform will not
help the bulk of older workers, who are under age 65 and thus not yet
eligible for Medicare. A growing consensus of experts agrees that “the
employer-sponsored system of bene ts in its current form may not be
sustainable, largely because productivity growth is unlikely to support
rising bene t costs” (Government Accountability Of ce 2006a, p. 26).
Health insurance reform of a more comprehensive nature, as addressed
in detail by Swartz (2008), is needed to remove this serious barrier to
employment for other older workers.
Over the coming decades, work for pay is likely to be increasingly
important to the nancial well-being of many older Americans. Higher
rates of employment at older ages also could help with addressing the
long-term funding problems faced by the Social Security and Medicare
To date, public policy initiatives intended to increase employment
among older Americans have focused primarily on monetary incentives.
Recent changes to the Social Security system that allow Social Security
recipients to earn more without having their bene ts reduced, that make
the present value of bene ts roughly independent of the age of retire-
ment rather than favoring those who retire early, and that raise the age
at which recipients qualify for full bene ts should make employment
tb08fogjch5.indd 193tb08fogjch5.indd 193 9/10/2008 12:47:15 PM9/10/2008 12:47:15 PM
194 Abraham and Houseman
signi cantly more attractive to older Americans. Furthermore, trends in
private-sector bene ts should reinforce the incentives associated with
these recent changes in the Social Security system. In particular, the
marked shift from de ned-bene t pension plans to less generous de-
ned-contribution pension plans and the sharp declines in the coverage
of retiree health insurance plans, together with the fact that savings for
retirement have not risen to offset the reduction in the generosity of the
pension and health bene ts available to retirees, should provide addi-
tional encouragement for Americans to work at older ages.
Although the altered nancial incentives associated with recent
policy and labor market changes unquestionably will be important in
promoting employment among older Americans, we have argued in this
chapter that policies designed to make work more attractive nancially
should be accompanied by policies designed to improve the function-
ing of labor markets for older workers. Our research as well as research
by others points to the special challenges to remaining employed that
older individuals face, even when they possess the ability and the desire
to continue working. These problems are particularly acute for low-
educated workers and for older individuals who attempt to transition to
new jobs.
We have advocated policies to help ensure that older workers who
wish to remain employed are able to do so. The measures we propose
should make it easier for older workers to search for a job and should
help to address some of the legitimate concerns that employers have
about hiring older workers. Although these measures are unlikely to
address fully the problems we have diagnosed, they would make an
excellent start and seem likely to have a signi cant societal payoff.
This paper was prepared for A Future of Good Jobs: America’s Challenge in the Global
Economy, a conference sponsored by the W.E. Upjohn Institute held in Washington,
DC on June 22, 2007. We have bene ted from conversations with many individuals,
including Linda Levine, Blake Naughton, Debra Whitman, and Julie Whittaker of the
Congressional Research Service; Dianne Blank, Barbara Bovbjerg, Mindy Bowman,
Alicia Cackley, and Sigurd Nilsen of the Government Accountability Of ce; and Em-
ily Allen, Dalmer Hoskins, Sara Rix, and Jim Seith of AARP. Richard Hobbie of the
National Association of State Workforce Agencies kindly assisted us with an informal
tb08fogjch5.indd 194tb08fogjch5.indd 194 9/10/2008 12:47:16 PM9/10/2008 12:47:16 PM
Removing Barriers to Work for Older Americans 195
survey of association members. We are also grateful to Lillian Vesic-Petrovic for excel-
lent research assistance.
1. For a description of these rules, see Deloitte (2006).
2. Although the share of older individuals who work should increase, the effect of
this change on total hours of work is ambiguous. Total hours of work could in-
crease if those choosing phased retirement would otherwise have quit working
altogether, or they could decrease if those choosing phased retirement would oth-
erwise have worked full time. Although the context is somewhat different, the
analysis reported by Gustman and Steinmeier (2007) of the effects of eliminating
the Social Security earnings test for those below normal retirement age suggests
the former effect may dominate.
3. This work updates research reported in Abraham and Houseman (2005). That pa-
per contains a more detailed discussion of the HRS and the questions pertaining to
future work and retirement plans that we use.
4. For instance, consider an individual who stated in the 1992 interview that he
planned to reduce his work hours by 1994. We count that individual as having re-
alized his stated plans if he was working at least eight fewer hours per week either
at the 1994 or at the 1996 interview. Because we only observe work hours at the
point in time of the interview, we do not know if an individual who is observed
to be working the same hours as before or who has stopped working altogether
reduced hours between interviews.
5. These estimates come from multivariate regression analyses that control for age,
gender, and time period in addition to education level. This analysis is available
from the authors on request.
6. We do not consider the Employment Service programs, which provide free labor
exchange services to job seekers and employers but do not offer the more inten-
sive counseling, job placement, or training services of the other federal programs
discussed in this chapter.
7. Figures for program year 2000 were affected by the transition from JTPA to WIA
and do not re ect true changes in the population served. More generally, the ac-
curacy of data on the number of exiters from JTPA and WIA programs has been
questioned because states have discretion in de ning whom to count as an exiter
(General Accounting Of ce 2004; Government Accountability Of ce 2005c). As
long as how states de ne exiters has not changed signi cantly over time, these
data should be indicative of broad trends both in the total number served and in
the number of older individuals served.
8. Public information campaigns and employer guidelines to combat discrimination
against older workers have been a leading strategy used in many countries to try
to increase the hiring of older workers (OECD 2006).
9. This gure comes from Table 5.3. For the purpose of the calculations reported
there we have assumed that all program exiters aged 55 and older were aged
tb08fogjch5.indd 195tb08fogjch5.indd 195 9/10/2008 12:47:16 PM9/10/2008 12:47:16 PM
196 Abraham and Houseman
AARP. 1998. Boomers Approaching Midlife: How Secure a Future? Wash-
ington, DC: AARP. Public Policy Institute.
econ/d16687_boomers.pdf (accessed September 12, 2007).
———. 2005. The Business Case for Workers Age 50+: Planning for To-
morrow’s Talent Needs in Today’s Competitive Environment. A Report for
AARP Prepared by Towers Perrin.
workers_ fty_plus.pdf (accessed August 8, 2007).
———. 2006. Business Executives’ Attitudes Toward the Aging Workforce:
Aware But Not Prepared? Report on Survey Conducted by BusinessWeek
Research Services for AARP. Rev. ed. Washington, DC: AARP. http://assets (accessed August 8, 2007).
Abraham, Katharine G., and Susan N. Houseman. 2005. “Work and Retirement
Plans among Older Americans.” In Reinventing the Retirement Paradigm,
Robert L. Clark and Olivia S. Mitchell, eds. New York and Oxford: Oxford
University Press, pp. 70–91.
Abraham, Katharine G., and James L. Medoff. 1984. “Length of Service and
Layoffs in Union and Nonunion Work Groups.” Industrial and Labor Rela-
tions Review 38(1): 87–97.
Arizona Mature Worker Initiative. 2006. Year One Outcomes and Rec-
ommendations. A Report to Governor Janet Napolitano. Phoenix: Ari-
zona Mature Worker Initiative.
SENIORWEBCASTSTARNS.PDF (accessed August 8, 2007).
Autor, David H., and Mark G. Duggan. 2006. “The Growth in the Social Se-
curity Disability Rolls: A Fiscal Crisis Unfolding.” Journal of Economic
Perspectives 20(3): 71–96.
Bendick, Marc Jr., Charles W. Jackson, and J. Horacio Romero.1996. “Em-
ployment Discrimination Against Older Workers: An Experimental Study
of Hiring Practices.” Journal of Aging and Social Policy 8(4): 25–46.
Buchmueller, Thomas, Richard W. Johnson, and Anthony T. Lo Sasso. 2006.
“Trends in Retiree Health Insurance, 1997–2003.” Health Affairs 25(6):
Burtless, Gary, and Joseph F. Quinn. 2002. “Is Working Longer the Answer for
an Aging Workforce?” Issue in Brief No. 11. Boston: Center for Retirement
Research at Boston College.
Chan, Sewin, and Ann Huff Stevens. 2001. “Job Loss and Employment Pat-
terns of Older Workers.” Journal of Labor Economics 19(2): 484–521.
Congressional Budget Of ce. 2003. Baby Boomers’ Retirement Prospects: An
Overview. Washington, DC: Congressional Budget Of ce. http://www.cbo
tb08fogjch5.indd 196tb08fogjch5.indd 196 9/10/2008 12:47:16 PM9/10/2008 12:47:16 PM
Removing Barriers to Work for Older Americans 197
.gov/ftpdocs/48xx/doc4863/11-26-BabyBoomers.pdf (accessed August 8,
Costa, Dora L. 1998. The Evolution of Retirement: An American Economic
History, 1880–1990. Chicago: University of Chicago Press.
Costo, Stephanie. 2006. “Trends in Retirement Plan Coverage Over the Last
Decade.” Monthly Labor Review 129(2): 58–64.
Deloitte. 2006. Securing Retirement: An Overview of the Pension Protection
Act of 2006. Washington, DC: Deloitte.
Dychtwald, Ken, Tamara J. Erickson, and Robert Morison. 2006. Workforce
Crisis: How to Beat the Shortage of Skills and Talent. Boston: Harvard
Business School Press.
Eibner, Christine, Alice Zawacki, and Elaine Zimmerman. 2007. “Older Work-
ers’ Access to Employer-Sponsored Retiree Health Insurance, 2000–2004.”
Center for Economic Studies Discussion Paper CES 07-12. Washington,
DC: U.S. Census Bureau, Center for Economic Studies.
Employment and Training Administration. 2006. “Common Measures Policy
for the Employment and Training Administration’s (ETA) Performance Ac-
countability System and Related Performance Issues.” Training and Em-
ployment Guidance Letter No. 17-05. Washington, DC: U.S. Department of
Labor, Employment and Training Administration.
Ernst and Young. 2006. The Aging of the U.S. Workforce: Employer Challeng-
es and Responses. New York: Ernst and Young.
Farber, Henry S. 2005. “What Do We Know about Job Loss in the United
States? Evidence from the Displaced Workers Survey, 1984–2004.” Work-
ing Paper No. 498. Princeton, NJ: Industrial Relations Section, Princeton
———. 2006. “Is the Company Man an Anachronism? Trends in Long Term
Employment in the U.S., 1973–2005.” Network on Transitions to Adult-
hood Research Network Working Paper. Philadelphia: Network on Transi-
tions to Adulthood.
20with%20cover%20sheet.pdf (accessed August 8, 2007).
———. 2007. “Job Loss and the Decline of Job Security in the United States.”
Working Paper No. 520. Princeton, NJ: Industrial Relations Section, Princ-
eton University.
Federal Interagency Forum on Aging-Related Statistics. 2006. Older Ameri-
cans: Key Indicators of Well-Being. Update 2006. Hyattsville, MD: Fed-
eral Interagency Forum on Aging-Related Statistics. http://www.agingstats.
gov/agingstatsdotnet/Main_Site/Data/2006_Documents/OA_2006.pdf (ac-
cessed September 17, 2007).
Freeman, Richard B. 2006. “Is a Great Labor Shortage Coming? Replacement
tb08fogjch5.indd 197tb08fogjch5.indd 197 9/10/2008 12:47:16 PM9/10/2008 12:47:16 PM
198 Abraham and Houseman
Demand in the Global Economy.” NBER Working Paper No. 12541. Cam-
bridge, MA: National Bureau of Economic Research.
Friedberg, Leora, and Anthony Webb. 2005. “Retirement and the Evolution of
Pension Structure.” Journal of Human Resources 40(2): 281–308.
Gale, William G., Jonathan Gruber, and Peter R. Orszag. 2006. “Improv-
ing Opportunities and Incentives for Saving by Middle- and Low-Income
Households.” Hamilton Project Discussion Paper 2006-2. Washington, DC:
Brookings Institution.
General Accounting Of ce. 2003. Older Workers: Employment Assistance
Focuses on Subsidized Jobs and Job Search, but Revised Performance
Measures Could Improve Access to Other Services. Report to the Ranking
Minority Member, Subcommittee on Employer-Employee Relations, Com-
mittee on Education and the Workforce, House of Representatives. GAO-
03-350. Washington, DC: U.S. General Accounting Of ce.
———. 2004. Workforce Investment Act: States and Local Areas Have Devel-
oped Strategies to Assess Performance, but Labor Could Do More to Help.
Report to Congressional Requesters. GAO-04-657. Washington, DC: U.S.
General Accounting Of ce.
Ghilarducci, Teresa, and Wei Sun. 2006. “How De ned Contribution Plans and
401(k)s Affect Employer Pension Costs.” Journal of Pension Economics
and Finance 5(2): 175–196.
Government Accountability Of ce. 2005a. Private Pensions: Information on
Cash Balance Pension Plans. Report to Congressional Requesters. GAO-
06-42. Washington, DC: U.S. Government Accountability Of ce.
———. 2005b. Older Workers: Labor Can Help Employers and Employees
Plan Better for the Future. Report to Congressional Requesters. GAO-06-
80. Washington, DC: U.S. Government Accountability Of ce.
———. 2005c. Workforce Investment Act: Labor and States Have Taken Ac-
tions to Improve Data Quality, but Additional Steps Are Needed. Report
to Congressional Requesters. GAO-06-82. Washington, DC: U.S. Govern-
ment Accountability Of ce.
———. 2006a. Employee Compensation: Employer Spending on Bene ts Has
Grown Faster Than Wages, Due Largely to Rising Costs for Health Insur-
ance and Retirement Bene ts. Report to Congressional Requesters. GAO-
06-285. Washington, DC: U.S. Government Accountability Of ce.
———. 2006b. “Senior Community Service Employment Program: Labor
Has Made Progress Implementing Older Americans Act Amendments of
2000, but Challenges Remain.” Testimony before the Special Committee
on Aging, U.S. Senate. GAO-06-549T. Washington, DC: U.S. Government
Accountability Of ce.
tb08fogjch5.indd 198tb08fogjch5.indd 198 9/10/2008 12:47:16 PM9/10/2008 12:47:16 PM
Removing Barriers to Work for Older Americans 199
———. 2006c. Trade Adjustment Assistance: Most Workers in Five Layoffs
Received Services, but Better Outreach Needed on New Bene ts. Report
to Congressional Requesters. GAO-06-43. Washington, DC: U.S. Govern-
ment Accountability Of ce.
Greenhouse, Steven, and Michael Barbaro. 2006.“Wal-Mart to Add Wage Caps
and Part-Timers.” New York Times, October 2, A:1.
Gustman, Alan L., and Thomas Steinmeier. 2007. “Projecting Behavioral Re-
sponses to the Next Generation of Retirement Policies.” NBER Working
Paper 12958. Cambridge, MA: National Bureau of Economic Research.
Hellerstein, Judith K., and David Neumark. 2007. “Production Function and
Wage Equation Estimation with Heterogeneous Labor: Evidence from a
New Matched Employer-Employee Data Set.” In Hard-to-Measure Goods
and Services: Essays in Honor of Zvi Griliches, Ernst R. Berndt and Charles
M. Hulten, eds. Chicago: University of Chicago Press, pp. 31–71.
Helman, Ruth, Craig Copeland, and Jack VanDerhei. 2006. “Will More of Us
Be Working Forever? The 2006 Retirement Con dence Survey.” EBRI Is-
sue Brief No. 292. Washington, DC: Employee Bene t Research Institute.
Hutchens, Robert. 2007. “Worker Characteristics, Job Characteristics, and Op-
portunities for Phased Retirement.” IZA Discussion Paper No. 2564. Bonn,
Germany: Institute for the Study of Labor.
Johnson, Richard W. 2007. “What Happens to Health Bene ts after Retire-
ment?” Issue in Brief. Work Opportunities for Older Americans, Series 7.
Boston: Center for Retirement Research at Boston College.
Judy, Richard W., and Carol D’Amico. 1997. Workforce 2020: Work and Work-
ers in the 21st Century. Indianapolis: Hudson Institute.
Lahey, Joanna. 2008. “Age, Women and Hiring: An Experimental Study.”
Journal of Human Resources 43(1): 30–56.
Leonhardt, David. 2007. “One Safety Net Is Disappearing. What Will Fol-
low?” New York Times, April 4, C:1.
Levine, Linda. 2007. Older Displaced Workers in the Context of an Aging and
Slowly Growing Population. CRS Report for Congress RL33054. Washing-
ton, DC: Congressional Research Service.
Madrian, Brigitte C., and Dennis F. Shea. 2001. “The Power of Suggestion:
Inertia in 401(k) Participation and Savings Behavior.” Quarterly Journal of
Economics 116(4): 1149–1187.
McCormack, Lauren A., Jon R. Gabel, Nancy D. Berkman, Heidi Whitmore,
Kay Bailey Hutchison, Wayne L. Anderson, Jeremy Pickreign, and Nathan
West. 2002. “Retiree Health Insurance: Recent Trends and Tomorrow’s
Prospects.” Health Care Financing Review 23(3): 17–34.
Medoff, James L., and Katharine G. Abraham. 1980. “Experience, Perfor-
mance, and Earnings.” Quarterly Journal of Economics 95(4): 703–736.
tb08fogjch5.indd 199tb08fogjch5.indd 199 9/10/2008 12:47:17 PM9/10/2008 12:47:17 PM
200 Abraham and Houseman
———. 1981. “Are Those Paid More Really More Productive? The Case of
Experience.” Journal of Human Resources 16(2): 186–216.
Munnell, Alicia H., Kevin E. Cahill, and Natalia A. Jivan. 2003. “How Has
the Shift to 401(k)s Affected the Retirement Age?” Center for Retirement
Research Issue in Brief No. 13. Boston: Center for Retirement Research at
Boston College.
Munnell, Alicia H., and Steven A. Sass. 2007. “The Labor Supply of Older
Americans.” Paper prepared for the Federal Reserve Bank of Boston sum-
mer conference “Labor Supply in the New Century,” held in Boston, June
National Governors Association. 2007. “NGA Center Eyes Increasing Em-
ployment and Volunteer Opportunities for Retirees: Eight States In-
vited to Policy Academy on Civic Engagement.” News release, May
24. Washington, DC: National Governors Association. http://www.nga
?vgnextoid=9b3a084726fb2110VgnVCM1000001a01010aRCRD (ac-
cessed August 8, 2007).
OECD. 2006. Live Longer, Work Longer. Ageing and Employment Policies
Project. Paris: Organisation for Economic Co-operation and Development.
Purcell, Patrick. 2005. “Older Workers: Employment and Retirement Trends.”
In Reinventing the Retirement Paradigm, Robert L. Clark and Olivia S.
Mitchell, eds. New York and Oxford: Oxford University Press, pp. 55–70.
Quinn, Joseph F. 1999. “Has the Early Retirement Trend Reversed?” Paper
presented at the rst annual joint conference of the Social Security Admin-
istration’s Retirement Research Consortia, held in Washington, DC, May
Scholz, John Karl, Ananth Seshadri, and Surachai Khitatrakun. 2006. “Are
Americans Saving ‘Optimally’ for Retirement?” Journal of Political Econ-
omy 114(4): 607–643.
Scott, Frank A., Mark C. Berger, and John E. Garen. 1995. “Do Health Insur-
ance and Pension Costs Reduce the Job Opportunities of Older Workers?”
Industrial and Labor Relations Review 48(4): 775–791.
Stevens, Ann Huff. 2006. “The More Things Change, the More They Stay
the Same: Trends in Long-Term Employment in the United States, 1969–
2002.” NBER Working Paper 11878. Cambridge, MA: National Bureau of
Economic Research.
Swartz, Katherine. 2008. “Revising Employers’ Role in Sponsoring and Fi-
nancing Health Insurance and Medical Care.” In A Future of Good Jobs?
America’s Challenge in the Global Economy, Timothy J. Bartik and Susan
N. Houseman, eds. Kalamazoo, MI: W.E. Upjohn Institute for Employment
Research, pp. 81–118.
tb08fogjch5.indd 200tb08fogjch5.indd 200 9/10/2008 12:47:17 PM9/10/2008 12:47:17 PM
Removing Barriers to Work for Older Americans 201
Technical Panel on Assumptions and Methods. 2003. Report to the Social Secu-
rity Advisory Board. Washington, DC: Social Security Advisory Board. http:// (accessed August 8, 2007).
Thompson, Lawrence H. 2004. “Social Security Reform and Bene t Adequa-
cy,” Retirement Project Brief No. 17. Washington, DC: Urban Institute.
U.S. Census Bureau. 2002. “Projections of the Total Resident Population by
5-Year Age Groups and Sex, with Special Age Categories: Middle Series,
1999–2010.” Summary Table NP-T3, National Population Projections.
Washington, DC: U.S. Census Bureau.
tb08fogjch5.indd 201tb08fogjch5.indd 201 9/10/2008 12:47:17 PM9/10/2008 12:47:17 PM
tb08fogjch5.indd 202tb08fogjch5.indd 202 9/10/2008 12:47:17 PM9/10/2008 12:47:17 PM
A Future of Good Jobs?
America’s Challenge
in the Global Economy
Timothy J. Bartik
Susan N. Houseman
W.E. Upjohn Institute for Employment Research
Kalamazoo, Michigan
Bartik & Houseman.indb 5 2/29/2008 1:42:00 PM
Library of Congress Cataloging-in-Publication Data
A future of good jobs? America’s challenge in the global economy / Timothy J. Bartik,
Susan N. Houseman, editors.
p. cm.
The outgrowth of a conference sponsored by the Upjohn Institute in Washington,
D.C., in June 2007. Includes bibliographical references and index.
ISBN-13: 978-0-88099-331-9 (pbk. : alk. paper)
ISBN-10: 0-88099-331-6 (pbk. : alk. paper)
ISBN-13: 978-0-88099-332-6 (hardcover : alk. paper)
ISBN-10: 0-88099-332-4 (hardcover : alk. paper)
1. Labor market—United States—Congresses. 2. Labor supply—United
States—Congresses. 3. Manpower policy—United States—Congresses. 4. Industrial
management—United States—Congresses. 5. Occupational training—United
States—Congresses. I. Bartik, Timothy J. II. Houseman, Susan N., 1956- III. W.E.
Upjohn Institute for Employment Research.
HD5724.F88 2008
© 2008
W.E. Upjohn Institute for Employment Research
300 S. Westnedge Avenue
Kalamazoo, Michigan 49007-4686
The facts presented in this study and the observations and viewpoints expressed are
the sole responsibility of the authors. They do not necessarily represent positions of
the W.E. Upjohn Institute for Employment Research.
Cover design by Alcorn Publication Design.
Index prepared by Diane Worden.
Printed in the United States of America.
Printed on recycled paper.
Bartik & Houseman.indb 6 2/29/2008 1:42:00 PM
... The demographic shift among workers also has reinforced companies' decisions to limit their fi nancial obligations for retiree health benefi ts. (Abraham and Houseman [2008] have a more detailed discussion in Chapter 5 of how retiree health benefi ts are likely to change in the next decade.) Since 1988, the fraction of companies with 200 or more employees that are offering retiree health insurance has declined sharply, from 66 percent to 33-35 percent in 2005 and 2006 . ...
The share of workers who are self-employed rises markedly with age. Given policy concerns about inadequate retirement savings, especially among those with lower education, and the resulting interest in encouraging employment at older ages, it is important to understand the role that self-employment arrangements play in facilitating work among seniors. New data from a survey module fielded on a Gallup telephone survey distinguish independent contractor work from other self-employment and provide information on informal and online platform work. The Gallup data show that, especially after accounting for individuals who are miscoded as employees, self-employment is even more prevalent at older ages than suggested by existing data. Work as an independent contractor is the most common type of self-employment. Roughly one-quarter of independent contractors aged 50 and older work for a former employer. At older ages, self-employment generally – and work as an independent contractor specifically – is more common among the highly educated, accounting for much of the difference in employment rates across education groups. We provide suggestive evidence that differences in opportunities for independent contractor work play an important role in the lower employment rates of less-educated older adults.
This chapter presents an overview of some of the most important topics arising from the confluence between labour economics and ageing, and focuses on the labour market participation of older people. It sets out the scope of labour economics with respect to individual ageing and alternative views and definition of ‘older’ workers. The chapter also raises aspects of supply and demand, paid employment, unemployment, job search and scarring effects, voluntary work, and inactivity—including a focus on retirement.
Deindustrialization in the United States has triggered record-setting joblessness in manufacturing centers from Detroit to Baltimore. At the same time, global competition and technological change have actually stimulated both new businesses and new jobs. The jury is still out, however, on how many of these positions represent a significant source of long-term job quality and security. Where Are All the Good Jobs Going? addresses the most pressing questions for today's workers: whether the U.S. labor market can still produce jobs with good pay and benefits for the majority of workers and whether these jobs can remain stable over time. What constitutes a 'good' job, who gets them, and are they becoming more or less secure? Where Are All the Good Jobs Going? examines U.S. job quality and volatility from the perspectives of both workers and employers. The authors analyze the Longitudinal Employer Household Dynamics (LEHD) data compiled by the U.S. Census Bureau, and the book covers data for twelve states during twelve years, 1992-2003, resulting in an unprecedented examination of workers and firms in several industries over time. Counter to conventional wisdom, the authors find that good jobs are not disappearing, but their character and location have changed. The market produces fewer good jobs in manufacturing and more in professional services and finance. Not surprisingly, the best jobs with the highest pay still go to the most educated workers. The most vulnerable workers-older, low-income, and low-skilled-work in the most insecure environments where they can be easily downsized or displaced by a fickle labor market. A higher federal minimum wage and increased unionization can contribute to the creation of well paying jobs. So can economic strategies that help smaller metropolitan areas support new businesses. These efforts, however, must function in tandem with policies that prepare workers for available positions, such as improving general educational attainment and providing career education. Where Are All the Good Jobs Going? makes clear that future policies will need to address not only how to produce good jobs but how to produce good workers. This cohesive study takes the necessary first steps with a sensible approach to the needs of workers and the firms that hire them.
Purpose of the Study: This study involved an in-depth exploration of the employment barriers of older Korean immigrants in Los Angeles. This qualitative study used data obtained from 6 focus groups and 5 individual interviews. Participants were 36 older Korean immigrants living in Los Angeles, aged 50 years and older, and either unemployed or employed in part-time or full-time work. A grounded theory analytical approach and constant comparison method were used. Ten major themes emerged as employment barriers for older Korean immigrants and were categorized as stereotype, human capital, and acculturation barriers. Ageism among employers specific to Korean culture, lack of English proficiency, separation from U.S. culture, marginalization from both Korean and U.S. cultures, and lack of social networks were important themes. In addition, older Korean immigrants experienced multiple interconnected barriers. The findings highlight the importance of using a multidimensional approach to explore employment barriers among older Korean immigrants who face multiple obstacles in finding jobs. Implications for local governments and Korean communities and potential services to support employment opportunities for older Korean immigrants are discussed.
Full-text available
Telework is a workplace arrangement in which employees have some degree of flexibility in work location and hours. The term 'Telework' was first coined in the 1970's to describe situations where information and communication technologies were used to support work activities undertaken away form the traditional office-based workplace. The subsequent three decades have seen many published reports on the issues surrounding Telework adoption and use. However much of this research has only examined the advantages and disadvantages of Telework use and has not adopted broader research perspectives to examine the deeper issues and the roles played by the various affected stakeholder groups. Over the same period, the incidence of Telework has increased significantly. The motivation for this paper is to develop a conceptual model capable of providing clear direction for research into the adoption and use of Telework. We then examine the usefulness of this model by identifying and framing a series of research projects aimed at addressing some of the existing gaps in the literature concerning Telework impacts.
From 1992-93 to 2005, there was an overall drop in retirement coverage; participation in defined contribution plans eclipsed that in defined benefit plans, and the features of retirement plans changed in tandem with the declining participation.
We solve each household's optimal saving decisions using a life cycle model that incorporates uncertain lifetimes, uninsurable earnings and medical expenses, progressive taxation, government transfers, and pension and social security benefits. With optimal decision rules, we compare, household by household, wealth predictions from the life cycle model using a nationally representative sample. We find, making use of household-specific earnings histories, that the model accounts for more than 80 percent of the 1992 cross-sectional variation in wealth. Fewer than 20 percent of households have less wealth than their optimal targets, and the wealth deficit of those who are undersaving is generally small.
This paper uses a telephone survey of 950 employers to examine employer-side restrictions on phased retirement. The survey not only collected information on establishment level policies, but it also asked questions about a specific worker's opportunity for phased retirement. The paper uses these data to first establish that employers are selective when offering opportunities for phased retirement. It then examines what worker and job characteristics are particularly important in the selection process.
Winner of the 1998 Paul A. Samuelson Award given by TIAA-CREF, The Evolution of Retirement is the first comprehensive economic history of retirement in America. With life expectancies steadily increasing, the retirement rate of men over age 64 has risen drastically. Dora L. Costa looks at factors underlying this increase and shows the dramatic implications of her findings for both the general public and the U.S. government. Using statistical, and demographic concepts, Costa sheds light on such important topics as rising incomes and retirement, work and disease, the job prospects of older workers, living arrangements of the elderly, the development of a retirement lifestyle, and pensions and politics. "[Costa's] major contribution is to show that, even without Social Security and Medicare, retirement would have expanded dramatically."—Robert J. Samuelson, New Republic "An important book on a topic which has become popular with historians and is of major significance to politicians and economists."—Margaret Walsh, Business History
This chapter begins by describing the aging of the US population and it summarizes historical data on older workers' labour force participation. It then considers information on older persons' employment and receipt of pension income, which are discussed in the context information on the proportion of workers who claim retired-worker benefits before the full retirement age (65 years and 4 months for people who reach age 65 in 2004). Recent proposals to promote phased retirement through amendments to sections of the Internal Revenue Code (IRC) that govern the taxation of pension income are discussed.
As the members of the “baby boom” generation—people born between 1946 and 1964—approach retirement, the demographic profile of the U.S. workforce will undergo a substantial shift as a large number of older workers will be joined by relatively few new entrants to the labor force. According to the Census Bureau, there will be 204 million Americans aged 25 or older in 2010. By 2030, this number will increase by 23% to more than 251 million. Most of this growth will occur among people aged 65 and older. The Census Bureau estimates that while the number of people between the ages of 25 and 64 will increase by 15.5 million (9.4%) between 2010 and 2030, the number of people aged 65 and older is projected to grow by 31.7 million, or 79.2%. Labor force participation begins to fall after age 55. In 2008, 91% of men and 76% of women aged 25 to 54 participated in the labor force. In contrast, just 70% of men and 59% of women aged 55 to 64 were either working or looking for work in 2008. Labor force participation among persons aged 55 and older is influenced by general economic conditions, eligibility for Social Security benefits, the availability of health insurance, and the prevalence and design of employer-sponsored pensions. For example, labor force participation among people 55 and older may increase due to the trend away from defined-benefit pension plans that offer a monthly annuity for life to defined contribution plans that typically pay a lump-sum benefit. The declining percentage of employers that offer retiree health insurance also may result in more people continuing to work until they are eligible for Medicare at 65. Census Bureau data show that the percentage of men and women aged 62 and older who work in paid employment has risen over the past several years. In March 2009, 52% of men aged 62 to 64 were employed, compared with 42% in 1990 and 47% in 2000. Of men aged 65 to 69, 33% were employed in March 2009, compared with 26% in 1990 and 30% in 2000. Among women 62 to 64 years old, 41% were working in March 2009, compared with 28% in 1990 and 35% in 2000. Among women 65 to 69 years old, 25% were working in March 2009, compared with 17% in 1990 and 20% in 2000. There also has been a trend toward more full-time employment among older Americans who work. In March 2009, 69% of employed men aged 65 to 69 were working full-time, compared with 56% in 1990 and 61% in 2000. Among working women aged 65 to 69, 54% worked full-time in March 2009, compared with 44% in both 1990 and 2000. As more workers reach retirement age, employers may try to induce some of them to remain on the job, perhaps on a part-time basis. This is sometimes referred to as “phased retirement.” Several approaches to phased retirement—job sharing, reduced work schedules, and rehiring retired workers on a part-time or temporary basis—can be accommodated under current law. The Pension Protection Act of 2006 (P.L. 109-280) allows pension plans to begin paying benefits to workers who have not yet separated from their employers at the earlier of age 62 or the pension plan’s normal retirement age, which in most plans is 65. Some employers would like to be able to pay partial pension distributions to workers who have reached the pension plan’s early retirement age, even if it is earlier than age 62. This would require a change in federal law.
[Exerpt] As the baby boom generation anticipates retirement, a growing proportion of older Americans are in fact remaining in the workforce. Labor force participation rates for older women have increased significantly since the mid-1980s, and for older men, since the mid-1990s, according to an updated report from the government's Federal Forum on Aging-Related Statistics. The labor force statistics are among several updated facts and figures in the Forum's databook series on aging.
This article uses data from the Health and Retirement Study to examine the employment patterns of workers aged 50 and above who have experienced an involuntary job loss. Hazard Models for returning to work and for exiting post displacement employment are estimated and used to examine work patterns for 10 years following a job loss. Our findings show that a job loss results in large and lasting effects on future employment probabilities. Four years after job losses at age 55, the employment rate of displaced workers remains 20 percentage points below the employment rate of similar nondisplaced workers. Copyright 2001 by University of Chicago Press.