Content uploaded by Julia Lane
Author content
All content in this area was uploaded by Julia Lane
Content may be subject to copyright.
Industrial & Labor Relations Review
|Number 4Volume 59 Article 82
2006
Moving Up or Moving On: Who Advances in the
Low-Wage Labor Market?
Fredrik Andersson
Harry J. Holzer
Julia I. Lane
Review of Moving Up or Moving On: Who Advances in the Low-Wage Labor Market?, by Fredrik
Andersson, Harry J. Holzer, and Julia I. Lane. Industrial & Labor Relations Review, Vol. 59, No. 4.
Available at: http://digitalcommons.ilr.cornell.edu/ilrreview/vol59/iss4/82
668 INDUSTRIAL AND LABOR RELATIONS REVIEW
benefit public employee pensions (Gary Ander-
son and Keith Brainard). With defined contri-
bution plans not only placing more risk on
workers but also giving them more responsibil-
ity for decision-making, financial education is
playing an increasingly important role in assur-
ing that the pension system meets its objectives.
Pension accounting rules are probably the area
least well understood by most economists who
study pensions, so the chapter on accounting
provides a valuable service by explaining the
rules and their importance in affecting the in-
vestment decisions of pension plan sponsors.
The chapter on defined benefit public employee
pensions presents a picture in sharp contrast
with conditions in the private sector, because
the dramatic decline in private sector defined
benefit plan coverage has not occurred in the
public sector. The reasons for that difference
are not fully explored.
The international chapters discuss pension
developments in Canada (Silvana Pozzebon),
Japan (Masaharu Usuki), and Sweden (Annika
Sundén). Of these three retirement income
systems, Canada’s is closest to that of the United
States. Canada is experiencing a decline in
defined benefit plans and a move toward de-
fined contribution plans, though at a much
slower rate than the United States. Japan has a
system of voluntary carve-outs, whereby employ-
ers may voluntarily withdraw their workers from
part of social security if the employers provide a
defined benefit plan that meets a minimum
standard of generosity. This system has encoun-
tered some of the problems that critics have
claimed would result from President George W.
Bush’s proposal for voluntary carve-out indi-
vidual accounts as part of U.S. Social Security
reform. In the past few years, there has been a
rapid decline in the number of workers partici-
pating in the Japanese carve-out system. Swe-
den, by contrast, has mandatory individual ac-
counts with a contribution rate of 2.5% of pay
that are an add-on to a generous base social
security system that is financed by a contribu-
tion rate of 16% of pay. The designers of the
Swedish system tried to learn from the experi-
ences of the mandatory individual account sys-
tems in Latin America, particularly regarding
ways to minimize the system’s administrative
costs. The experiences of all three countries
may provide valuable lessons for U.S. policy-
makers as they consider various reform options
for both Social Security and private pensions.
The Organization for Economic Coopera-
tion and Development (OECD), whose mem-
bership includes the United States and other
high-income countries, favors workers postpon-
ing retirement as the main way for societies to
deal with population aging and longer life ex-
pectancy. That policy raises questions as to the
ability of people in different occupations to
work at older ages, the changing nature of the
physical and psychological demands of work,
the effects of work on health, the demand for
older workers, and the effect of older work-
ers’ continued work on the demand for
younger workers. Those issues, which are not
covered by Reinventing the Retirement Paradigm,
are sufficiently complex to warrant a book of
their own.
The book’s editors and many of the chapter
authors are leading scholars or practitioners in
the pension field. Across a broad range of
topics, the book contributes to the advance of
our understanding of pension and retirement
issues. It is well worth reading for both the
novice and the seasoned specialist.
John A. Turner
Senior Policy Advisor
AARP Public Policy Institute
Human Resources,
Management, and Personnel
Moving Up or Moving On: Who Advances in
the Low-Wage Labor Market? By Fredrik
Andersson, Harry J. Holzer, and Julia I.
Lane. New York: Russell Sage Foundation,
2005. 192 pp. ISBN 0-87154-057-6, $29.95
(cloth).
Although it joins an already vast literature on
how to improve the labor market success of the
American poor, Moving Up or Moving On is un-
usual in using rich data on workers and their
employers to suggest that the labor market suc-
cess of a poor individual is heavily influenced by
his or her employer. Research frequently con-
siders how the labor market success of the poor
is influenced by their human capital character-
istics, such as education level and training pro-
gram participation. Whether because of data
limitations or lack of interest, however, this
research rarely considers more than a few em-
ployer characteristics, such as firm size and
industry. Because of the unusual data used by
Andersson, Holzer, and Lane (hereafter AHL),
they can consider how the labor market ad-
vancement of a poor individual depends on
many more employer characteristics, such as
BOOK REVIEWS 669
whether the employer pays a wage premium or
has previously helped the poor advance.
The authors use the Census Bureau’s Longi-
tudinal Employer-Household Dynamics (LEHD)
database, which is comprised of unemployment
insurance wage records from various states,
linked to Census data. The book includes LEHD
data on five states: California, Florida, Illinois,
Maryland, and North Carolina. The records for
these states span nine years (1993–2001), allow-
ing analysis of fairly long-term effects.
The LEHD data include information on quar-
terly earnings of 96% of private non-farm em-
ployees, for a yearly total, in the five states
studied, of 37 million workers and 1.8 million
employers (Table 2.3). By linking to Census
data, the LEHD also includes information on
the age, gender, race, and ethnicity of 96% of
these individuals. Some limitations of the LEHD
data are that we only know workers’ quarterly
earnings, not work hours, and for most workers
we do not know education or family character-
istics.
AHL’s analysis focuses on what happens over
time to “prime-age” workers with consistent la-
bor force attachment who initially have low
earnings. In this study, prime-age workers are
defined as those who were aged 25–54 in 1993;
workers with consistent labor force attachment
are those who had at least one quarter of earn-
ings greater than $2,000 in each year from 1993
through 2001; and initially low-earnings work-
ers are those whose annual earnings, in 1998
dollars, were less than $12,000 in each year from
1993 through 1995. These restrictions reduce
the sample for most of the analyses from 37
million to 1.4 million individuals (p. 153). AHL
examine whether these individuals significantly
increased their earnings during three-year peri-
ods subsequent to 1993–95, particularly the years
1999–2001. As one would expect, they find that
earnings increased more for younger white males
than for other demographic groups.
The innovative part of the analysis is the
demonstration that labor market success for
initial low earners was linked to characteristics
of individuals’ subsequent employers. Consis-
tent with previous studies, these researchers
find that labor market success was more fre-
quent for those who were hired in construction
and manufacturing, as well as for those hired by
larger firms. But AHL turn up other correlates
of success, as well:
• moving on to an employer different from the pri-
mary 1993–95 employer, particularly if the individual
moved to a new employer but then stayed to accumu-
late job tenure;
• working during 1999–2001 for an employer that
had low average employee turnover;
• working for a temporary employment agency dur-
ing 1996–98, and then moving to another employer
during 1999–2001; and
• working for an employer that, over the years 1993–
98, paid a high “fixed (over time) earnings pre-
mium,” as indicated by an estimate that controls for
fixed (over time) personal effects of individual work-
ers and each worker’s labor market experience.
Another noteworthy finding is that hiring of
low-earnings workers was concentrated in rela-
tively few firms. For example, 1,000 employers
in these five states did one-quarter of all hires of
these initially low-earnings workers.
In addition, workers’ success after being hired
was concentrated in relatively few employers.
AHL define a worker as “fully escaping” low-
earnings status if the worker’s earnings exceeded
$15,000 during each year of a three-year period.
Such escapes occurred disproportionately at a
few companies. For example, of the 6,000 health
service employers in Maryland, fewer than 2,000
hired an initially low-earnings worker, and just
20 of these employers accounted for one-quar-
ter of all low-earnings workers hired by that
industry who “fully escaped” low-earnings status
(p. 113).
The authors predicate several policy recom-
mendations on their empirical findings. Two of
the options they suggest are expanding the use
of labor market intermediaries, including tem-
porary agencies, to place low-earnings workers
in better jobs, and providing subsidies for cus-
tomized training in order to encourage more
employers to provide advancement opportuni-
ties for low-earnings workers.
The three authors should be commended for
boldly stretching a rich database to its limits to
address policy issues. Also praiseworthy is their
frank admission of our uncertainty about how
employers might influence their employees’
labor market success. Recognizing that an
employer’s personnel practices are not perfectly
predictable based on the employer’s size and
industry, they develop a methodology that al-
lows for “unobserved” employer-specific factors
to influence the labor market success of initially
low-earnings workers. Such an analysis could
only be done using data over time for most
employees of many employers, which would
seem to require the LEHD data for the United
States.
However, I think the authors could have
done more to identify limitations of the study
arising from the way they capture these “em-
ployer effects.” (The only such treatments I
670 INDUSTRIAL AND LABOR RELATIONS REVIEW
find are two brief discussions on pages 144 and
152.) Because so many of the analyses consider
how outcomes are correlated with employer
“fixed effects” on workers’ earnings, the book
would have been stronger with an additional
appendix detailing how these earnings effects
are estimated, and discussing how these estima-
tion procedures constrain the bounds within
which the effects can be interpreted.
Part of the issue is that the estimation of
employer effects and person effects on earnings
is only identified by the persons who switch
employer. With only a few years of data for each
person, there is substantial measurement error
in estimating person fixed effects. In addition,
the specification used to estimate person and
employer fixed effects on earnings levels does
not allow for person and employer effects on
earnings growth, which are plausible. As a
result, what is estimated to be an employer fixed
effect on earnings levels may in part reflect the
kinds of workers the employer hires. For ex-
ample, an employer may tend to hire workers
whose true person fixed effects on earnings
levels are higher than these workers’ measured
person fixed effects, or workers whose earnings
growth trends are greater than the average
person’s growth rate trend.
The policy discussion also could have more
fully explored some of the limitations of what
can be inferred from these estimates. Even if
the estimates do reflect causal effects of em-
ployer characteristics, the average effects of
employer characteristics on workers’ success
under current labor market policy need not
reflect the marginal effects under policies that
change how workers are placed in jobs. For
example, in research completed after this book
was written, David Autor and Susan Houseman
found that welfare clients, if they are essentially
randomly assigned to welfare-to-work agencies
that tend to place more welfare clients in tem-
porary jobs, have lower earnings later on (“Do
Temporary Help Jobs Improve Labor Market
Outcomes for Low-Skilled Workers?” Working
Paper 11743, National Bureau of Economic
Research, November 2005). Autor and
Houseman’s results may not contradict the re-
sults reported in AHL’s book. Low-earnings
workers who are currently placed in temporary
jobs may on average experience a long-run ben-
efit, but the marginal additional individuals
placed in temporary jobs by policy variations
may not.
However, these are modest caveats regarding
a unique book that should spark needed future
research. Andersson, Holzer, and Lane persua-
sively argue that employers matter in the suc-
cess of low-earnings workers. As the book points
out, this finding should lead to more research
to figure out the specific mechanisms that ex-
plain why and how employers matter. And it
should lead to exploring a wide variety of poli-
cies to see if we can alter the distribution of
opportunities offered by employers to low-earn-
ings workers. Some of these policies might fail,
but continued efforts to work with employers
are likely to be fruitful for both research and
policy regarding low-earnings workers.
Timothy J. Bartik
Senior Economist
W.E. Upjohn Institute for
Employment Research
The Mismanagement of Talent: Employability
and Jobs in the Knowledge Economy. By Phillip
Brown and Anthony Hesketh. New York:
Oxford University Press, 2004. x, 278 pp.
ISBN 0-19-92-6953-X, $39.95 (paper); 978-
0-19-92-6954-9, $124.50 (cloth).
A 1997 landmark study by McKinsey & Com-
pany exposed the “war for talent” as a strategic
business challenge and key driver of company
performance. Although the McKinsey consult-
ants were certainly not the first to note the shift
toward a knowledge-based economy (KBE)—
Peter Drucker predicted the rise of the knowl-
edge worker over a half-century ago—their re-
port has perhaps done most to promulgate the
“war for talent” perspective and the view that
companies that are victorious on this battlefield
will be those that realize competitive advantage
in the KBE.
In The Mismanagement of Talent, Brown and
Hesketh argue that rooted within the dominant
discourse of the “war for talent” are several core
assumptions that have shaped our perspective
on employability in the KBE. The most central
of these is that there is a limited pool of talent
capable of rising to senior managerial posi-
tions, which creates fierce competition to re-
cruit the best and brightest. The perception of
talent as a limited commodity is seen as driving
organizations to diversify their talent pools and
adopt more rigorous recruitment and selection
tools in an effort to get the right people, with
the right knowledge, into the right jobs. That
perception also shapes public policy, as govern-
ments focus on expanding access to higher
education and dismantling barriers to talent
