Content uploaded by Peter. Cappelli
Author content
All content in this area was uploaded by Peter. Cappelli on Jul 01, 2018
Content may be subject to copyright.
ILR Review, 68(2), March 2015, pp. 251 –290
DOI: 10.1177/0019793914564961. © The Author(s) 2015
Journal website: ilr.sagepub.com
Reprints and permissions: sagepub.com/journalsPermissions.nav
SKILL GAPS, SKILL SHORTAGES, AND SKILL
MISMATCHES: EVIDENCE AND ARGUMENTS FOR THE
UNITED STATES
PETER H. CAPPELLI*
Concerns over the supply of skills in the U.S. labor force, especially
education-related skills, have exploded in recent years with a series
of reports not only from employer-associated organizations but also
from independent and even government sources making similar
claims. These complaints about skills are driving much of the debate
around labor force and education policy, yet they have not been
examined carefully. In this article, the author assesses the range of
these charges as well as other evidence about skills in the labor force.
Very little evidence is consistent with the complaints about a skills
shortage, and a wide range of evidence suggests the complaints are
not warranted. Indeed, a reasonable conclusion is that overeduca-
tion remains the persistent and even growing condition of the U.S.
labor force with respect to skills. The author considers three possi-
ble explanations for the employer complaints and the associated
policy implications.
Assertions of widespread problems with the supply of skill in the U.S.
labor force have been common in recent years. Stories are frequent in
the popular press of individual employers who report that they cannot ll
vacancies,1 but a large number of detailed reports have also been produced
by business associations, individual companies, and independent organiza-
tions arguing that skill problems are widespread. The rise of these stories is
especially surprising because they appear to have increased since the 2008
Great Recession, when the ood of unemployed job seekers—most of them
recently employed—far exceeded available job opportunities (Bureau of
Labor Statistics [BLS] 2014b).
*P H. C is Professor of Management at the Wharton School and Director of the Center
for Human Resources and Research Associate at the National Bureau of Economic Research (NBER).
Thanks to colleagues at the World Economic Forum’s Global Agenda Council on Employment for help-
ing me understand developments outside the United States; to David Coats for help with the UK litera-
ture; to Pascaline Descy and colleagues at the European Center for the Development of Vocational
Training (CEDEFOP); and to Jerry Jacobs, Paul Harrington, and Paul Osterman for helpful comments.
The information used in this article is accessible through the references in the bibliography.
564961ILRXXX10.1177/0019793914564961ILR ReviewSkill Gaps, Shortages, and Mismatches
research-article2015
K: labor markets, lateral hiring, job skills, education, skill mismatch, skill gap
1A Google search for the phrase skill gap showed more than 330,000 references just in 2013.
252 ILR REVIEW
The evidence driving the complaints about skills does not necessarily
appear in a way that labor market experts might expect to see it, such as in
rising wages. Instead, it comes directly from employers, who report, typically
in surveys, difculties hiring the kind of workers they need. The assertions
explaining their reported difculties center on the idea that the academic
achievement of high school leavers is inadequate or that not enough col-
lege graduates are in practical elds such as computer science and engi-
neering. The recommendations from these reports include increased
immigration and use of foreign workers as well as efforts to shape the majors
that college students choose.
These reports have had a powerful inuence on public debate about the
state of the labor market as well as the performance of high schools and col-
leges. Virtually all of them are framed in terms of concerns about the econ-
omy as a whole, but the fact that the producers of many of these reports
have a material interest in the outcomes of the policies they are attempting
to inuence is difcult to escape.
The arguments in this article examine the claims associated with complaints
about the supply of skills in the United States. The evidence I present here as
well as evidence from other sources suggests little merit to these claims. An
important question raised by these reports, however, is who is responsible for
the skill level of job candidates and, ultimately, of employees. Traditional
employer human resource practices, such as assessing the abilities of job appli-
cants, training employees for current jobs, and developing them internally for
future roles saw employers as responsible for securing the skills they needed.
The thrust of these reports suggests that the education sector, especially
public-funded education, and the job candidates themselves should be respon-
sible for producing the skills that employers want. Such a change would have
profound implications for society and is worth considering seriously.
Framing the Problem
The arguments that problems exist with the supply of skills available in the
labor market take various forms. The most extreme complaint is the idea that
widespread shortfalls have been found in the basic skills of future employees.
The cause is usually attributed to the failure of the education system, espe-
cially K-12 public education, to provide students with these basic skills. We
refer to that position as a skills gap, following its use in policy discussions.
A second complaint focuses more on job-related skills of the kind associ-
ated with particular occupations, such as the common assertion that the
United States is short on engineers or information technology (IT) special-
ists. We refer to this assertion as a skills shortage.
The nal concern, which is much more commonly articulated outside
the United States, is the more general idea that at any given time the supply
of skills and the demand for skills could be out of synch in either direction:
oversupply or undersupply. This situation could occur in specic labor mar-
kets, although with respect to educational credentials it is usually
253SKILL GAPS, SHORTAGES, AND MISMATCHES
considered at the country level. We refer to it as a skills mismatch. A skill
shortage is obviously a particular type of skill mismatch, and a skills gap
could be a general form of mismatch. All these complaints collectively can
be referred to as skill problems.
The rst challenge in assessing the assertions about skills problems is to
have a conceptual framework for understanding the relationship between
workers and their skills against employer needs. One approach, tradition-
ally associated with the topic of internal labor markets and the academic
eld of human resources (and, before that, personnel), suggests that match-
ing skills to job requirements is an employer problem. Over time, employers
have internalized the supply of labor, selecting for general abilities at entry-
level positions and then training and developing employees over a working
lifetime to meet their specic skill needs (Jacoby 1983). That approach
appears to have eroded substantially in recent years (e.g., Cappelli 1999), an
issue we return to later in the article.
The other approach focuses on the labor market as the mechanism for
meeting employers’ skill requirements. The idea of job matching between
employers and job seekers implies a job search, which typically assumes that
employers have job requirements that are generally determined exogenously
and that employers then go searching for job applicants who have those
skills. The search process is realistically described as two-sided if both employ-
ers and employees are looking for a match, and a good match is one in which
the skills of the applicants and the requirements of the job t closely, neither
a shortfall nor an oversupply of skills relative to those requirements.
In typical economics models of job search, the process is reasonably pas-
sive. Employers make offers to job applicants, who accept the job when the
offer matches or beats their reservation price. Employers raise wages to
attract better applicants whose attributes are closer ts for those job require-
ments, and they lower wages if an excess supply of such applicants exists. The
notion of a shortage is foreign to this model and to most all economics-based
models. Indeed, shortages in general are typically seen as occurring only in
the context of market failure, such as wartime wage freezes or restrictions on
mobility, and to be temporary until candidates and employer adjust.2
In practice, of course, employers can search more extensively through
recruiting activities and more careful selection, and applicants can search
actively by securing better information about vacancies. We also know that
job requirements are not exogenous from the supply of applicants; a short-
fall of applicants that leads to higher wages causes employers, in turn, to
substitute capital for labor to create new jobs with lower skill requirements.
Empirical evidence indicates that employers also lower the skill require-
ments for given jobs when labor is relatively scarce and raise them when
higher-quality applicants are plentiful (Walsh 1977; Brencˇicˇ 2010).
2See Mortenson (1986) for a basic framework; Rogerson, Shimer, and Wright (2005) for an overview
of job search models; and Borjas (2010, Chap. 6) for a survey of general labor supply questions.
254 ILR REVIEW
What is less clear in the typical models is how the supply of skills affects
employer decisions on production systems and, ultimately, productivity.
Certainly, “better” workers, who are absent less, who shirk less, and who
work harder, will improve organizational productivity and performance,
even if nothing changes about their jobs (e.g., see Cascio 2008 for a review).
But whether more skilled applicants per se cause employers to innovate, to
adopt more effective practices, or to change the way that jobs are performed
is an open question that is often part of the skill gap arguments.
The recent assertions about skill problems have quite a different underlying
model in mind, although it is typically unstated, and it does not include a role
either for internal labor markets or for the labor market. Instead, the argu-
ments are akin to input–output models associated with operations research
optimization exercises. Perhaps the closest analogy is with supply chain models
where suppliers are trying to produce just the right amount of output to meet
the needs of their clients at the previously agreed price (see, e.g., Cachon and
Terwiesch 2006). Skills are seen as coming with the applicant to the job, and job
requirements are absolute, such that candidates either have the necessary skills
to do a job or not and, if not, they cannot do the job. Finally, an important goal
for public education in these arguments, including public colleges and univer-
sities, is seen as providing graduates that employers would like to hire.
Reports of Skill Gap and Skill Shortages
The Skill Gap Idea
The broadest and perhaps most general complaint about skills has been the
skill gap idea, that some systematic shortfall exists in skills, broadly dened,
across entire age cohorts of the population. Typically the argument is that
the decline is associated with the poor skills of school leavers, and the expla-
nation for that shortfall is usually that schools have failed, so academic per-
formance of students has declined.
Concerns such as this in the United States trace their contemporary roots
to the Cold War and the 1958 National Defense Education Act, which
increased funding for science and engineering education in an effort to
compete with the Soviet Union. The A Nation at Risk (National Commission
on Excellence in Education 1983) report highlighted declines in student
achievement of all kinds in the 1970s and helped cement in the public mind
for decades afterward that U.S. schools were failing.
An equally attention-getting report by the Carnegie-funded National
Center on Education and the Economy, America’s Choice: High Skills or Low
Wages? (1990), argued explicitly that productivity growth depended on
increasing the skill level of the U.S. workforce. Although the report esti-
mated that only 5% of employers thought they faced any skill problems, that
low number was attributed to the fact that employers were not introducing
new, high-performance work practices that would increase productivity.
Among its recommendations were the establishment of national educa-
tional standards for students based on international standards, the
255SKILL GAPS, SHORTAGES, AND MISMATCHES
establishment of skill certications, and national training boards to orga-
nize smooth transitions between school and work.
A separate report for the U.S. Secretary of Labor represented an exten-
sion of the America’s Choice study. The Secretary’s Commission on Achiev-
ing Necessary Skills ([SCANS] 1991) also envisioned a future in which
employee empowerment had increased, workplaces had moved toward high-
performance work systems that required greater skills, and employers trained
constantly and saw human resources as investments rather than costs. The
recommendations, which became known as SCANS Skills, were central to
the policy debates of the 1990s. They called for a generic set of skills from
high schools that included basic skills (reading, writing, math, etc.); thinking
skills, such as decision making; and personal attributes such as responsibility.
Developed in the Republican George H. W. Bush administration, the SCANS
ideas were nevertheless embraced by the Democratic Clinton administration
and dominated the discussion of skills throughout the 1990s.
The most important action on skills in the 1990s was arguably the school-
to-work movement, based in part on SCANS arguments, which asserted that
the way to improve student skills and to increase employability was to bring
schools and employers closer together in an effort to smooth the transition
from school into work. In practice that meant apprenticeships, coop pro-
grams, internships, and other arrangements that would help students see
the practical value of classroom lessons, rst, by using more business and
workplace examples in the classroom and, second, by seeing how those
examples could be applied at work. The School-to-Work Opportunities Act
of 1994 (STWO) provided administrative and nancial support to help
build those connections. (See Stull and Sanders 2003 for an overview of the
school-to-work movement and public policy.)
Joyce and Neumark (2001) found that 64% of schools had at least one
school-to-work program with employers (the most popular of which was
“job shadowing”), and data from the National Longitudinal Survey of Youth
suggest that 38% of students participated in school-to-work programs. Cen-
sus data from establishments found 71% of for-prot establishments
reported that they were involved in some school-to-work program with their
local schools (Cappelli 2001). Because establishments represent the loca-
tion where business takes place rather than the rm per se, this suggests
quite extensive business involvement.
The high-water mark of the SCANS effort was the National Skills Stan-
dards Act of 1994, which was designed to create a voluntary system of
national standards for job skills, overseen by a National Skills Standards
Board. The initial efforts were organized around industries, but consensus
on standards proved elusive, and interest waned. The board’s last publica-
tion, seeking comments on an approach for creating standards, was in 2002.
The Labor Shortage Argument
With the sunset of the STWO Act in 1999, a new administration less inter-
ested in a role for government in affairs related to business, and the
256 ILR REVIEW
recession of 2001 that cooled off the tight labor market, the debate about
skills changed sharply. Funding for SCANS-related efforts and school-to-
work programs ended, as did arguments about the need to push employers
toward high-performance work systems and to create skill standards.
Most of the skills-related arguments following the 2001 recession came
initially from consultants who asserted that, in the near future, enough peo-
ple to meet the labor demand would simply not be available. Those argu-
ments were kicked off by McKinsey & Co.’s “The War for Talent” study
(Chambers et al. 1998), which observed the smaller “baby bust” age cohort
born in the 1970s and asserted that this would soon cause a shortfall in
middle-age employee talent, although why a special need for middle-age talent
would appear was not clear. Similar reports followed, such as one from the
U.S. Chamber of Commerce (2006: 13), which assumed that the impending
retirement of the baby boom cohort would lead to an absolute decline in the
size of the labor force and a “severe worker shortage.” (See Public Policy and
Aging Report 2004 for a collection of articles arguing that a labor shortage is
coming, and Challenger 2003 and Carnevale 2005 for similar arguments.)3
These arguments are puzzling in part because the basic claim that the
population and potential labor force is or will be shrinking seems to be a
simple misreading of the facts: Only the rate of increase in the labor force
was expected to slow, assuming baby boomers did not delay retirement (see
Congressional Budget Ofce [CBO] 2011 for a review of forecasts).4
Several researchers pointed out other problems with these labor shortage
arguments, such as that the anticipated slowdown in the rate of increase in
the labor force is trivial compared to changes in the demand for labor asso-
ciated with the business cycle (Cappelli 2003; Freeman 2006; Neumark,
Johnson, and Mejia 2011; Harrington and Sum, forthcoming). Despite the
obvious problems with these labor shortage arguments, the Society of
Human Resource Management (SHRM 2003) reported that large numbers
of employers in the early 2000s were preparing for a labor shortage pre-
dicted to occur by 2010. None of these projections proved to be right.
Skill Gaps and Science, Technology, Engineering,
and Math Skills Reports
If the SCANS approach was aspirational (“here’s what the workforce should
have”), the more recent arguments have been grounded in the present,
3Concerns that U.S. demographics will cause some shortfall in the supply of labor, affecting business
and the economy, continued to be voiced by the Aspen Institute (2003), Dychtwald, Erickson, and Mor-
rison (2006), and others. The Great Recession and unemployment rates at double-digit levels did not
stop these arguments. Venneberg and Eversole (2010), Bluestone and Melnik (2010), and others asserted
that the retirement of the baby boomers will lead to apocalyptic shortages, with as many as 30 to 40% of
jobs in sectors such as health care being vacant.
4The fact that the labor-force participation rate in the United States has declined since the Great
Recession in 2008 may say little about the longer-term availability of workers because much of the decline
appears to be driven by the lack of job opportunities. In March 2014, for example, 2.2 million individuals
marginally attached to the labor force were available for work but not counted as unemployed because
they were not actively searching for jobs. In March 2008, just before the economic downturn began, the
gure was 1.4 million, according to the Bureau of Labor Statistics Employment Situation News Release
for those months. See BLS (2013) for information on the labor-force participation rates.
257SKILL GAPS, SHORTAGES, AND MISMATCHES
asserting that currently a shortfall exists in the skills of the workforce. The
American Society of Training and Development (ASTD), whose members
hold jobs as trainers and in employee development, for example, asserted
that view in a series of annual reports beginning in 2003. The most interest-
ing part of their most recent report (ASTD 2012) is that the most important
explanations members saw as the causes of the shortfall in skills were man-
agement actions internal to the organization.5 Skills problems were seen as
self-inicted by management.
Other reports argued that a shortage of skills will be associated with col-
lege education. The President’s Council on Jobs and Competitiveness, a
business-led council (20 of its 24 members are from business), claimed that
the United States would have a shortfall of 1.5 million college graduates by
2020, citing McKinsey & Co. as the source (Jobs Council 2012a, 2012b; Pres-
ident’s Council on Jobs and Competitiveness 2012). In perhaps the most
alarmist report, Carnevale, Smith, and Strohl (2010) concluded that the
demand for college graduates in the United States will fall short of supply by
3 million individuals by 2018. Harrington and Sum (2010) pointed out that
one fundamental problem with projections such as these is the assumption
that every job currently held by a college graduate requires the skills associ-
ated with that degree. As Levine (2013) pointed out, 60% of parking-lot
attendants in Wisconsin have at least some college education, yet those jobs
surely do not require college-level skills.
The focus of skill gap arguments in the 2000s shifted from broad SCANS-
type skills to academic skills associated with science, technology, engineer-
ing, and math (STEM) education, particularly four-year college degrees.
Complaints from the business community about shortfalls in the supply of
such skills have been voiced before (e.g., Atkinson 1990 predicted a short-
fall of 400,000 scientists), but the intensity of the arguments increased
sharply, especially when they were joined to the debate about immigration.
A report from the U.S. Department of Commerce (1997) argued that a
severe shortage of IT workers existed in the United States, which required
increasing immigration as well as expanding education. That claim and the
report itself was criticized by a Government Accounting Ofce report (1998)
that noted methodological problems with it, such as the assumption that
only graduates with IT degrees can do IT-related jobs and the reliance on
evidence from a tech employer group (e.g., with a survey response rate of
only 14%). The National Research Council’s (2001) report on the IT work-
force avoided taking a clear position on whether an IT labor shortage
existed, but its recommendations relied heavily on traditional actions that
employers should take to address their perceived shortages, such as more
extensive training.
5The most common response was that their organization had changed “strategy, goals, markets, or
business models” such that the current workforce did not have the skills to meet those new approaches;
in second place was the lack of “bench strength,” or possible replacements for the current leadership;
and in third place was that cuts in training investments and lack of commitment to development caused
a shortfall. Only in fourth place was there a shortfall in the skills of applicants.
258 ILR REVIEW
The attention on skill problems shifted to STEM skills with the joint publica-
tion by the National Academies of Science, Engineering, and the Institute of
Medicine of “Rising above the Gathering Storm: Energizing and Employing
America for a Brighter Economic Future” (2007). The charge given to the com-
mittee was how best to support science and technology in the United States,
both in research and in business applications. A central argument of the report
was that the cost and availability of skilled labor in science and technology drives
multinational companies in their decisions about where to operate and that
increasing the supply of STEM graduates and allowing more foreign STEM
workers and foreign STEM students to enter and stay in the United States would
benet the industry. This view represented a sharp break from the America’s
Choice and SCANS idea that an important goal of policy was to raise wages.
Here the notion was that higher wages were actually a problem for the technol-
ogy industry and therefore for the advancement of technology.
Since then, arguments about skill shortages and STEM skills, in particu-
lar, have been commonplace. The National Academy of Sciences produced
six separate reports related to STEM skill issues just in 2012, many about
expanding the supply of skill. The President’s Council of Advisors on Sci-
ence and Technology produced one of many studies asserting a shortfall of
STEM grads and arguing that the United States needed an additional 1 mil-
lion such graduates to meet demand (President’s Council of Advisors on
Science and Technology 2012). These arguments about a coming labor
shortage are very similar to those made a decade earlier, the difference
being the focus on particular skills.
The Institute of Electrical and Electronics Engineers, which represents
engineers, disputed the claims of a shortfall of engineers, in particular, not-
ing a range of evidence about the difculty that many engineers had in get-
ting jobs and suggesting that actually a surplus of STEM graduates existed
relative to demand and that a large percentage of STEM jobs were not being
performed by individuals with degrees in that eld; few computer program-
mers, for example, have a B.S. in computer science (Charette 2013). Salz-
man, Kuehn, and Lowell (2013) also argued against the STEM shortage
idea in a report for the Economic Policy Institute (EPI). The strongest evi-
dence in their study arguably is that roughly half of recent engineering
graduates did not take jobs as engineers, and of those who do not, roughly
30% said no job was available for them and another 30% said that the terms
and conditions for those jobs were below market levels.6
6For graduates in IT, perhaps the hottest of the true STEM job markets, only about one-third failed
to take IT jobs. But about 30% of those said that the reason was that no job was available for them, and a
much larger percentage, 53%, said that the terms and conditions for such jobs were below the prospects
elsewhere. The extremes were represented by health care graduates (who are not always counted as
STEM graduates), almost three-quarters of whom got health care jobs, and all other STEM grads (e.g.,
science and math majors), only about 22% of whom got jobs in their eld. Atkinson and Stewart (2013)
published a rebuttal to the Salzman et al. (2013) report for The Information Technology and Innovation
Foundation (ITIF). Perhaps the most interesting point in their report is the observation that the spon-
sors of the various reports matter: EPI’s goal, they argued, is higher wages for tech workers, whereas their
goal was to promote economic growth, presumably by making tech labor easier and cheaper for IT
employers to engage. (EPI is sponsored in part by labor unions; ITIF is sponsored by tech employers.)
259SKILL GAPS, SHORTAGES, AND MISMATCHES
The Computer and Technology Industry Association (2012) produced an
unusually detailed report, again based on a survey of employers, in which
93% of employers responded said that they had a skills gap. Yet 90% also
responded that they were at least “moderately close” to “where they want to
be” with respect to skills, only 15% said that a factor in their skill problem
was insufcient focus on STEM education, and only 20% reported that the
problem was a limited pool of skilled IT workers, the essence of the STEM
skill shortage argument. A large part of their perceived skill gaps had to do
with “soft skills”—work ethic and motivation (almost 20% said that their
concern was only with soft skills, and about half reported that their concern
was equally divided between hard and soft skills).
Individual employers have also produced reports claiming that nation-
wide skills problems exist. For example, the health science company Bayer
issued a report (Bayer Corporation 2013) based on a survey of recruiters at
Fortune 1000 companies reporting that their companies were creating more
jobs for STEM graduates than for those with any other credentials, a surpris-
ing nding given that the BLS’s Occupational Outlook forecast (2014a)
projected that only two (registered and licensed nurses) of the top 20 occu-
pations with the most new openings through 2020 will require any postsec-
ondary education, let alone STEM degrees.7
A related complaint one often hears in STEM arguments is that U.S. col-
lege students do not want to take the hard majors in college, such as those
associated with STEM degrees, and instead are coasting in easier majors and
ones that have poor job prospects such as the liberal arts. In fact, the most
popular majors in four-year colleges seem to be the most vocational: business
is by far the most popular, with one in ve students majoring in it (365,000
per year) and in which job skills are the focus; education is the second most
popular major (104,000 per year), another vocational eld with a very close
tie to jobs. The National Center for Education Statistics ([NCES] 2013)
dened liberal arts rather narrowly, and only 47,000 students chose that
major, in contrast to 76,000 who were in engineering. Since 2000 to 2001,
STEM degrees have increased at a rate equal to or greater than the 32%
increase in overall bachelor degrees: engineering, up 31%; physical sciences,
up 34%; math/statistics, up 42%; and biology, up 50% (ibid., Table 313).8
More recent evidence suggests that entering college students have been
responding to the employer complaints about shortfalls in STEM graduates.
Jacobs and Sax (2014) found that student choices for majors in college have
shifted sharply, from 21% in STEM elds in 2007 to 28% in 2011. The ratio
of graduate degrees to bachelor degrees issued also rose from roughly one
out of three to one out of two from 1971 to 2012, and more than 85% of
those degrees are in applied elds (NCES 2014a: Table 318).
7Barnow, Trutko, and Schede Piatak’s (2013) investigation of occupation-specic and regional labor
shortages is an exception to those reviewed here in that it placed labor market adjustments and better
information front and center as the means for addressing such shortfalls.
8If we dene liberal arts more broadly to include humanities and the social sciences, the percentage
of such graduates declined from 40% of all bachelor degrees in 1971 to about one-third in 2012. Thanks
to Jerry Jacobs for this gure and for the graduate degree information as well.
260 ILR REVIEW
Consultant Reports
Virtually every major consulting company has also weighed in on either the
skills gap or skill shortage idea. As previously noted, Deloitte and the Manufac-
turing Institute (2011) argued that a massive shortage of qualied candidates
had been found in that industry; in addition, McKinsey Global Institute (2012)
argued that a shortage of tens of millions of educated workers across the devel-
oped economies would occur in the coming decades, and Pricewaterhouse
Coopers ([PwC] 2012) noted that chief executive ofcers (CEOs) believe that
labor shortages will impact their companies’ success in the near future.
Two consulting companies produced reports pointing in the opposite
direction, however. The Boston Consulting Group (2013) argued that little
evidence of tight labor markets for manufacturing workers exists because
wages are not rising. Accenture (2012) took a very different approach by sur-
veying workers rather than employers. They found, among other things, that
only one in ve workers had learned any new job skills through employer-
provided training in the previous ve years and that a plurality said that their
employer relied only on their formal education and prior job experience in
assigning tasks and jobs, not using any other assessments of their knowledge,
skills, and abilities. As a possible sign of changing norms, 68% of respon-
dents believed strongly that keeping their skills up to date in their current
job was their responsibility, not their employer’s (only 9% disagreed).
Perhaps the most inuential and widely quoted of the consultant reports
on skill problems have been those published by the Manpower Group, the
stafng rm, in large measure because their reports provide reasonably
detailed information from employers across countries on their perceived
skill problems. The limitations of the reports’ data are similar to those for
virtually all the business-based reports that gather data from employers: The
sampling frame used for the survey is not identied, but the inference is
strong that it is not representative of the population of employers from
which it is drawn (e.g., the sample is likely to be clients of Manpower); and
we do not know the sample size, the response rate to the survey, how the
questionnaire was structured, who the respondents were, and so on. Many
of the results also suggest problems of face validity. For example, in 2010,
14% of U.S. employers reported that they had difculty lling jobs, but a
year later in the 2011, when the economy and the labor market had barely
changed, the gure jumped to 52%. Then in 2013, when the economy had
begun to show some signs of improvement and one would expect hiring to
become more difcult, the percentage of employers reporting difculty
actually fell to 39% (Manpower 2013: 7). In addition, the wording of the
question asking about skill problems is not clear, which makes the validity of
the responses impossible to determine.9
9The tables in the report show responses to the question, “How much difculty are you having lling
jobs due to lack of available talent?” which is a question about degree. But the results are reported as the
percentage of employers having difculty lling jobs, which is a binary, yes/no question. Some places in
the text of the report refer to the same results as representing “hard-to-ll jobs.”
261SKILL GAPS, SHORTAGES, AND MISMATCHES
Nevertheless, the reports may be useful in providing some details about
what employers may mean when they say that they have difculty hiring. For
example, on the respondents’ list of the 10 most difcult jobs for U.S.
employers to ll are laborers, a job that requires no discernable skill; drivers
and production operators (factory workers), two positions that require min-
imal skill; and secretaries/administrative assistants, jobs that do not neces-
sarily require more than a high school education. That the difculties in
hiring for these jobs represent a real shortfall of individuals who are able to
perform the tasks required in those jobs is difcult to believe, especially
given the high rate of unemployment during this period. They may well
reect problems with hiring practices.
When asked what the shortfall was in the applicants they saw for their jobs,
almost 20% said that the job seekers were not willing to accept positions at
the rate of pay being offered, yet only 5% reported that they were planning
to raise the pay rate to deal with the difculties in hiring. Among the causes
that reected directly the attributes of the candidates, 31% of employers said
the shortfall was lack of experience, 6% reported that the problem was over-
qualication, and 8% said that applicants did not want the jobs because they
were contingent or part-time. Only a third of respondents reported that a
lack of hard skills, the category related to skills that could be learned in
school, was a problem they saw (Manpower 2013). These responses are hard
to square with the skills gap view or even the skills shortage argument.
Nothing in the employer reports indicates that difculties in hiring refer
specically to applicants who are recent school leavers, although the con-
clusion is typically that they are. The vast majority of the labor force is obvi-
ously not composed of recent school leavers. The BLS calculates that about
1.3 million recent school leavers in the age group 16 to 24 are in the labor
force (Spleen 2013), less than 1% of the U.S. labor force. The vast majority
of job candidates, therefore, have been out of school for some time, and so
the vast majority of the job applicants should be out of school for some time
as well (applicants include those currently employed who are looking for
better jobs as well as the unemployed).
Perhaps the most relevant employer-based evidence on skill problems
comes from other studies that asked recruiters, rather than higher-level
executives, about specic shortfalls that they saw in new hires, especially
those who were recent school leavers. Several such reports have appeared,
at least since the 1980s, and the conclusions have been similar. The com-
plaints consistently focused on shortfalls in factors associated with conscien-
tiousness (see Cappelli 1995). More recent surveys reported the same
conclusions. For example, a 2009 survey by the Business Roundtable found
that the top seven most commonly seen decits were in those workplace
attitudes. The rst shortfall associated with academic skills was oral commu-
nication, in eighth place.10 A 2011 hiring-manager survey found that, of the
15 attributes listed as important for success in their organizations, only
10See http://businessroundtable.org/studies-and-reports/american-worker-survey-telebrieng/.
262 ILR REVIEW
communication skills were related to an academic subject (Career Advisory
Board 2011). Wolf, Aspin, Waite, and Ananiadou (2010) reached the same
conclusion for the United Kingdom; employer complaints about school
leavers were about attitudes rather than skills.
Comparative Evidence
Some insight into the situation in the United States might be gained from
what employers elsewhere are reporting about skills. Although, as we have
seen, the validity and reliability of the data may be suspect, a survey by Man-
power found U.S. employers to be fourth out of 41 countries in their com-
plaints about hiring the applicants they wanted, far above countries with
tight labor markets such as Singapore and Norway, where we might expect a
relative shortfall of applicants.
The economy, legal framework, and business community in the United
Kingdom is arguably the most similar to the United States among the world’s
larger economies, which might make a detailed comparison useful. The busi-
ness community there has also been very active in efforts to persuade the gen-
eral public and policymakers about the need to improve workforce skills.
Although some of the UK arguments mirror those in the United States, oth-
ers, and the evidence behind them, are quite different. For example, the
Institute of Director’s (2010) report argued that a signicant gap exists
between the skills employers need and those that school leavers have but that
the biggest gap they saw is in soft skills: leadership, sales, communication, and
customer service skills, in that order. This mirrors the data but not the conclu-
sions of the U.S. reports. The Engineering Employer’s Federation (2013)
reported that the industries experiencing the biggest skill problems were con-
struction and agriculture—not high-skill industries—although technical skills
were the biggest shortfall in those two industries. It called for an expansion of
apprenticeship programs as its most important policy recommendation.
The Federation of Small Businesses (n.d.) and the British Chambers of
Commerce (2011) also made a strong case that apprenticeships are the key
to addressing perceived skill problems. The latter noted that, although 54%
of employers surveyed said that apprenticeships were not relevant to their
businesses, 30% of those who took apprentices did so out of civic responsi-
bility, which is similar to what U.S. employers reported a decade earlier in
census data (Cappelli 2001). And although almost two-thirds of employers
said that they had some engagement in programs to help students make the
transition from school to work, 14% had no training budgets at all and pre-
sumably did no training. (The gures for Europe as a whole are even lower,
with about one-third of employers reporting that they did not provide train-
ing. The main reason employers across Europe gave for not training was
that they tried to hire workers with skills so that they did not need to train.
See Continuing Vocational Training Statistics [CVTS] 2013.)
The most inuential of the UK business groups is the Confederation of
British Industry (CBI), and its 2013 report on skills followed the model of
263SKILL GAPS, SHORTAGES, AND MISMATCHES
recent U.S. reports closely. Indeed, it began by referencing U.S. complaints
about skills, then talked about failing schools as the cause using declining
performance compared to international standards as evidence, and nally
focused on improved STEM skills, using the U.S. term, as the solution to
skill problems. The evidence in its employer survey does not support that
conclusion, however, because only 12% of employers reported having dif-
culty hiring college graduates with STEM skills, and the skills they were most
satised with were actually the hard STEM skills associated with IT. Despite
this, respondents reported that their priority for schools and college was to
develop more business-relevant coursework.
Contrary to its recommendations, the CBI study also found that work atti-
tudes and aptitudes were the rst concern of British employers who are hir-
ing, followed by the appropriateness of the candidate’s academic degree for
the job. Prior work experience, the number one concern for U.S. employ-
ers, was third on the list of the UK employers. The UK employers respond-
ing reported much greater involvement in education than did those in the
United States: 81% report they had some formal ties to schools, 70% had
apprenticeships, 41% offered internships, and 61% said that they were will-
ing to do more to help schools. The CBI report implied that employer-
provided training was even rarer than the British Chambers of Commerce
(2011) report asserted. The CBI report said that 59% of employers provided
training for employees, implying that 41% did not.
Taken as a whole, the evidence from UK employers is different from their
U.S. counterparts in acknowledging that behavioral skills are the main con-
cern, not academic skills, and in seeing apprenticeships as the preferred
solution, something that is rarely mentioned in the U.S. reports. (Recent
practice in the United Kingdom has been for the government to pay most of
the costs of apprenticeships, especially for younger school leavers.) More
important, despite the many similarities shared by the two countries and the
United Kingdom having essentially the same unemployment rate as the
United States at the time the survey was done (8.1% in the United Kingdom
compared to 8.2% in the United States), UK employers ranked fourth from
the bottom out of 41 countries in complaints about skill problems com-
pared to the U.S. employers, who ranked fourth from the top(Manpower
2013). Whatever is driving the U.S. complaints, it appears to be something
not found in the United Kingdom.
Academic Research on Skills
The reports and evidence discussed so far were not written by or for an aca-
demic audience, nor do they reect any of the prior research on questions
concerning skills. Contemporary academic research on skill levels compared
to employer needs arguably began with Berg’s (1970) assertion that much of
the investment in skills through training and education was not productive
but was actually driven by hiring requirements that served other needs, such
as legitimacy, and by the interests of students in going to college for reasons
264 ILR REVIEW
other than getting jobs, a precursor to later arguments about signaling in
labor markets. The research took its most important turn with Freeman’s
(1976) study of the sharp decline in the wage premium of college graduates
from 1969 to 1974 measured relative to the wages of high school graduates.
Although his conclusion did not hold up in later periods (e.g., Smith and
Welch 1978), others made similar arguments about education mismatches,
especially Duncan and Hoffman (1981), who pursued the issue at the level of
the individual worker. They examined the returns to education for individ-
ual workers relative to the requirements of their job and found that the
returns were considerably lower for those who were overeducated compared
to those whose education levels equaled the job requirements.
Their paper kicked off a stream of similar studies that continue today,
especially outside the United States. In recent studies, McGuinness and
Sloane (2011) found wage penalties for education mismatches in the form
of more skill, independent from education per se, for men than jobs
required and that job satisfaction was signicantly worse for those who were
overeducated and overskilled for their positions. Sutherland (2012), using
the recent UK Skills and Employment Survey, argued that overqualication
in skills and education is widespread in the United Kingdom and that, as a
result, increasing skills and education further was not a sensible policy.
Instead, policies should focus on increasing the demand for skills. Feld-
stead, Gallie, Green, and Inanc (2013) used the same data and found that
overqualication of workers rose from 1986 to 2006 but then fell from 2006
to 2012, during the Great Recession. The decline reected in part increases
in hiring requirements, which were the measure used for job requirements,
and the fact that training and learning times were shortened for jobs; appli-
cants were expected to come to jobs with more of the skills they needed and
not to require training.11 Mosca and Wright (2013) used micro-data from
UK college graduates and found that long-term negative effects, or “scar-
ring,” were associated with a worker starting out overskilled; workers who
start out underemployed tend to stay that way (in contrast to the earlier
argument by Sicherman and Galor [1990] that those who are overqualied
for their jobs may advance faster).
Researchers studying other countries found similar results. Ng (2003)
found that the overeducated earned less in Canada and that, in general,
education–job requirement mismatches were reasonably common there.
Badillo-Amador, Lopez Nicolás, and Vila (2012) used the European House-
hold Panel Survey to show that, for Spain, skill mismatches had big, negative
11Gottschalk and Hansen (2003) found almost the reverse relationship in the United States, where the
probability of being a graduate employed in a job that did not require a graduate education declined
between 1983 and 1996 as the labor market got tighter. The implication might be that, in a down market,
U.S. employers are more likely to hire overqualied applicants whereas UK employers in similar circum-
stances are inclined to raise job requirements so that graduates are not overqualied. Gottschalk and
Hansen dened graduate jobs in a very different manner, however, based on outcomes that make com-
parisons difcult: Is the job title, in fact, lled by graduates, and does a wage premium exist in it for
graduates?
265SKILL GAPS, SHORTAGES, AND MISMATCHES
effects on wages and job satisfaction. Mavromaras and McGuinness (2007)
reported similar results for Australia—a small wage premium was associated
with being undereducated and a wage penalty was associated with being
overeducated. Mavromaras et al. (2013) used the new Household Income
and Labour Dynamics (HILDA) data from Australia and concluded that
being overeducated and overskilled were not identical but that both had
negative effects on wage outcomes. Mavromaras et al. (2013) also used the
HILDA data and found evidence of long-term scarring—being overskilled
led to long-term, negative effects on wages. Baert, Cockx, and Verhaest
(2013) found similar results for Flemish youth who were overeducated in
their rst job. The exception to the general overskilled story seems to be
Italy, where Cainarc and Sgobbi (2012) used face-to-face interviews with
respondents about skills and concluded that substantial undereducation
relative to job requirements occurred there.
These results suggest that in these countries skill mismatches are a com-
mon problem and also that being overskilled is a much more widespread
problem than being underskilled. Employees benet somewhat from being
underskilled relative to the requirements of the job they are performing, in
that their wages are higher there than in jobs that were a better t for their
skills (i.e., where the skill requirements are lower). The costs to being over-
skilled are borne by employees in the form of wages that are lower than they
would have earned in jobs that made full use of their skills. In a context in
which employees are increasingly asked to pay for their own skills, asking
them to be overskilled, when they will not be compensated for that excess
skill, is a burden.
The literature specic to the United States on skill mismatch is sparse.
Most of it is in sociology and focuses on the more general question as to
whether skill requirements in the economy are rising. The consensus view is
that overall requirements may be trending upward in recent decades, albeit
very slightly, although sorting out the changing requirements in specic
jobs is more difcult than assessing changes in the distribution of workers
across jobs. The most important factor driving the changes appears to be
the decisions of individual employers as to how they organize work (see
NRC 2008 for an overview). The study that most directly addresses the topic
of mismatches is Vaisey (2006), who compared educational qualications to
the education requirements of jobs using the U.S. Department of Labor’s
O*NET job classication system to dene the job requirements. The study
showed that the average worker in the United States is overqualied for his
or her job and that the amount of that overqualication has been increas-
ing in recent years.
Liu and Grusky (2013) used an interesting approach to assess changes in
skill requirements across the workforce. They began with the jobs of Current
Population Survey (CPS) respondents from the 1970s to 2013 as representa-
tive of the distribution of jobs in the economy. They then used the U.S.
Department of Labor’s O*NET database on job requirements to capture the
266 ILR REVIEW
skills associated with each job.12 They found increases in skill requirements
over the last 40 or so years, some to do with changes in the distribution of
jobs and some to do with changes in skills within jobs. The increases were
modest: academic skills (analytic, quantitative, and verbal) up only 4%; com-
puter skill requirements up 8%, surprisingly small given the dramatic
increase in the use of computers; and, especially relevant for the STEM skills
debate, no increase in science and engineering skill requirements.
Whether employers are hurt by underskilled workers and helped by over-
skilled workers is a more difcult question to answer. Acemoglu and Zili-
botti (2001) examined country-level data among less developed countries
and found that average skill levels below those deemed necessary for appro-
priate technologies used in those countries were associated with signicantly
lower country productivity. These skill levels were associated with the most
basic skills, such as low-level literacy. Bennett and McGuinness (2009) sur-
veyed prior research on skill mismatches and rm performance and
reported mixed results: some studies with no effect and others with negative
effects. They found that for electronics rms in Northern Ireland, hiring
complaints were mainly in job searches for experienced candidates. The
effects of unlled vacancies, in particular, on performance were negative
and substantial but only appeared when the analyses controlled for rm-
selection effects. Those effects were attributable largely to high-performing
rms. Kampelmann and Ryxc (2012) did nd positive productivity effects
associated with overeducated workforces in Belgium, however.
This academic work has many limitations, of course, especially construct
validity and measurement issues. Most of these studies used education as a
proxy for the level of skill an individual has or a job requires. No doubt one
reason for this is the roots of the skill argument in returns to education, but
a more practical reason is that education levels are much easier to assess
than skill levels. A second major problem is the difculty in assessing the
skills that jobs require. The best studies use objective assessments of the job,
which requires an independent job analysis, but most ask the respondents
themselves either what is required to be hired into the job or the extent to
which they believe their own skills are being used.
Structural Change and Vacancy Arguments
A different set of arguments asserts that the U.S. economy or the labor mar-
ket has changed in ways that have altered the balance between the supply
and demand of skills. Lazear and Spletzer (2012) investigated the idea that
job losses associated with the Great Recession have been in different seg-
ments of the economy than those in which recent job growth has been.
They rejected that view, noting that the pattern indicates something more
12The measure used here took the government’s ofcial analyses of skills—one-third of the knowl-
edge, skills, and abilities trio of requirements—for each occupation in 1979 and then compared them to
more recent assessments from incumbents.
267SKILL GAPS, SHORTAGES, AND MISMATCHES
like a cyclical downturn, in which job losses and subsequent gains are follow-
ing a pattern similar to those we have seen in previous recessions.
Rothstein (2012) examined a wide range of claims about possible changes
in the structure of the economy and the labor market and rejected them all. In
particular, he noted the absence of any upward pressure on wages that might
be associated with a shortfall between the supply and demand of labor by
industry or occupation or even a tight labor market. The best explanation for
perceived difculties in hiring by employers, he suggested, would be declines
in the effort spent on recruiting. Davis, Faberman, and Haltiwanger (2012)
provided evidence that employer efforts to recruit applicants per vacancy did
indeed decline during the Great Recession and have not recovered.
A measure that might well reveal hiring problems is vacancy data. Speci-
cally, if the amount of time that positions remain vacant increases, this may
suggest something about difculty in hiring and, in turn, about the supply of
qualied applicants. One of the most puzzling claims made by many of the
skill problem advocates is the notion that the presence of vacancies in the
labor force is evidence that jobs cannot be lled. The President’s Council on
Jobs and Competitiveness (Jobs Council 2012b) made that claim, asserting
that vacancies per se reect the inability to nd qualied applicants rather
than the usual view that they reect only the fact that time is required to ll
vacancies: Post the job advertisement, collect applicants, process them, hire
someone, and then close down the advertisement. Deloitte and the Manufac-
turing Institute (2012) claimed that 600,000 good jobs in U.S. manufactur-
ing cannot be lled because of a lack of qualied applicants, an astonishing
gure given that the BLS (2014b) reported only 220,000 total vacancies in
manufacturing during the year the data for the 2012 report was collected.
Osterman and Weaver (2014) investigated the labor market context in man-
ufacturing using their own survey and found that two-thirds of employers
reported having no vacancies and only 25% had vacancies open long enough
to suggest a difculty in lling them. The most common of their respon-
dents’ self-reported explanations as to why lling those long-term vacancies
were difcult was that candidates lacked industry-specic skills (41%), and
the second most common explanation was that the wages they were offering
were not sufcient to attract candidates (11%).
Whether lling vacancies is taking longer now than in the past is an
important question. The Beveridge curve gets at that indirectly by capturing
the relationship between the unemployment rate and the number of job
openings as a proportion of the labor force. Jobs that stay open get counted
again in each time-series estimate, so a change in the length of time required
to ll jobs would cause an apparent outward shift in the curve. Develop-
ments that cause jobs to be lled faster would do the opposite and cause an
apparent inward shift in the curve.
Barnichon, Elsby, Hobijn, and Sahin (2012), using the JOLTS data, found
that the Beveridge curve did shift for the United States after the Great Recession
in 2009 and that the shift was caused by a decline in hires per vacancy expected
at the relevant level of unemployment. Many factors could account for the
268 ILR REVIEW
relative decline in hiring, such as greater hiring of the currently employed
(which keeps the unemployment rate from declining while creating vacancies
elsewhere) and a decline in lling vacancies from within (which expands the
vacancy rate), but this would be the strongest evidence so far that problems with
the labor supply that make hiring more difcult might exist. The authors con-
cluded that a mismatch of skills in the form of unemployed job seekers coming
from different industries than those in which the vacancies were is not part of
the explanation, however, and the list of possible explanations outside the labor
supply altogether is extensive (e.g., the greater uncertainty in the economy).
Davis, Faberman, and Haltiwanger (2013) used the same data and also
found fewer hires than expected in the period since recovery from the Great
Recession ofcially began. They also found evidence of considerable variation
across employers in their ability, or perhaps inclination, to ll vacancies.
Among their strongest ndings is that growing rms and those lling many
vacancies were more likely to ll a reported vacancy than were rms that were
not growing or lling as many positions. These results suggest that something
about the manner in which rms are recruiting and selecting candidates may
explain why vacancies last longer, as the Davis et al. (2012) study concluded.13
Skill-Biased Technological Change
Among the most cited evidence in the academic literature on skill problems
are ndings under the heading of skill-biased technological change (SBTC).
In brief, the argument is traced to Tinbergen’s (1974) idea that new tech-
nologies require more skilled workers and that, over time, the continual
introduction of new technology therefore leads to a continual demand for
more skills. Societal investments in education increase the supply of skilled
workers and keep wages for workers with higher skills from spiraling upward.
Katz and Murphy (1992) produced compelling evidence consistent with
the Tinbergen view by looking at the college wage premium (measured rela-
tive to the wages of high school graduates) over time; they found that the
college wage premium closely tracked the relative supply of college gradu-
ates. Deviations from the prediction were seen as resulting from changes in
technology, much as deviations in the orbits of planets are seen in astron-
omy as evidence of gravity from unseen heavenly bodies. The fact that the
wage premium for college educations was rising in the 1980s even as the
relative supply of college graduates rose was taken as evidence of a shift in
technology biasing demand toward more skilled or educated workers.14
13Another important nding from their study is that 27% of hires took place in establishments that did
not report having a vacancy, which suggests that relying on vacancy data to assess hiring may be less than
perfect.
14The gap between the average wages for high school and for college graduates grew sharply after
1980, driven both by increases in wages for college graduates (in the late 1990s) and the decline of real
wages for high school graduates (in the late 1980s). As Acemoglu and Autor (2012) noted, however, col-
lege wages have been reasonably at since about 2002, and some of the high wages measured for college
graduates may come from including those with graduate degrees in with the college wage premium; real
wages for those with graduate degrees continued to rise through the 2000s, but real wages for those with
only a bachelor’s degree did not.
269SKILL GAPS, SHORTAGES, AND MISMATCHES
Technology per se is not specied or measured directly in typical SBTC
studies. It is assumed to be an attribute of the economy that is ever increasing
and is often proxied by a simple time trend. But computer use of the kind
Krueger (1993) and Autor, Katz, and Krueger (1998) examined is a favorite
illustration of such technology, and it is associated with higher wages and
more educated workers. Autor, Levy, and Murnane (2003) looked directly at
the effect of computer use on skill requirements and found that it increased
higher-skill, nonroutine tasks while reducing lower-skill, nonroutine tasks,
also consistent with a SBTC view.15 Goldin and Katz (2008) extended the
SBTC argument and used it to explain wage patterns across U.S. history.
In contrast, other studies suggested limitations to the SBTC argument.
Autor, Katz, and Kearney (2008) looked at changes in the distribution of
wages and concluded that increasing employment in higher-skill jobs and
decreasing employment in lower-skill jobs in the 1980s do explain rising
wages in the former and falling wages in the latter. But in the 1990s, employ-
ment in low-skill jobs also increased. Acemoglu and Autor (2011, 2012)
extended these analyses to the present and found that the college wage pre-
mium remained relatively steady in the 2000s despite a slowdown in the
increase in the relative supply of college graduates compared to high school
graduates. The authors inferred that the increase in the relative demand for
college graduates therefore also slowed down. Technology, of course, is only
one factor that can affect the demand for college workers; others include
international trade, outsourcing, and consumer demand patterns. Beudry,
Green, and Sand (2013) concluded that the demand for higher-skill jobs
that require college degrees is actually declining (see also Autor 2010) and
that college grads are forced to look to jobs that require less skills. In the
process, they “bump” the applicants without a college degree, who then end
up with even-lower-skilled jobs or none at all.16
More generally, the notion that technology is an incessant force creating
demand for higher skills is contradicted by research in other elds. In soci-
ology during the period after Tinbergen’s work, the dominant view of tech-
nology and skills was that technology was often designed precisely to reduce
skill requirements. Technology associated with scientic management, such
as assembly lines, clearly did reduce the average skill requirements, increase
the supply of labor that could perform most jobs, and lower wages in the
process. It also reduced the control and discretionary effort that workers
could exercise in those jobs (e.g., Braverman 1974). Since then, organization-
specic studies of the introduction of technology, such as computer-assisted
design and manufacturing, have suggested strongly that employer choices
15The related notion that IT technology has the inexorable effect of creating job losses through
increases in productivity in industries and occupations where it is used (Brynjolfsson and McAfee 2011)
is a powerful idea, but the relationships among IT use, productivity, and job loss change considerably
over time (Acemoglu et al. 2014).
16Whether technology has also created polarization in the labor market, increasing jobs at the top and
bottom of the skill distribution while shrinking the middle (e.g., Autor et al. 2003), is also an important
question. Acemoglu and Autor (2011) found evidence that polarization continues into the recent period,
and Mishel, Shierholz, and Schmitt (2013) asserted that it does not.
270 ILR REVIEW
determine whether skill requirements rise or fall for different workers (e.g.,
Zicklin 1987; Keefe 1990). Acemoglu (2002) offered an intermediate view,
asserting that, although technological change, dened broadly, in recent
years does appear to be biased toward creating greater demand for skill/
education, it does not necessarily have to be that way. The outcome depends
on the relative prices of skilled labor and, therefore, the incentives for
employers to reduce their use of such skills.
Even though the existence of a substantial college wage premium seems
to be compelling evidence that college skills have substantial value, what we
conclude from that premium should be considered carefully. First, because
it represents the difference in average wages for all college graduates com-
pared to all high school graduates, the college wage premium is not neces-
sarily representative of the experience of new hires, nor is it necessarily
predictive of the future college premium.17 For example, the college pre-
mium appears to have declined considerably during the Great Recession,
falling from 69 to 63% between 2008 and 2011 (NCES 2014c: Table 438).
Second, we know that college graduates are different from high school
graduates in ways other than their education, such as their socioeconomic
status, their abilities, and other dispositions, and that those differences also
affect their earnings. Assuming that a typical high school graduate is identi-
cal to the average college graduate except for education, and that the former
would make the same wage as the latter if he or she had a college degree,
would be wrong. Yet that assertion is commonly made, at least implicitly.
Third, the college premium has also been inuenced by factors that have
nothing to do with the demand for college grads. The decline of unions, for
example, held down wages disproportionately for high school graduates
and increased the college premium. Assuming that the college premium
proxies the demand for college graduates should require controlling for
the factors that disproportionately affect high school wages.
Fourth, using education as a proxy for the “skill” that employers want
should be interpreted with caution as well, given that the extensive literature
on job analysis shows that the knowledge, skills, and abilities that are used in
jobs have, at best, only a partial overlap with what is taught in typical college
courses. As noted elsewhere, no evidence exists that skill requirements in the
United States have risen at anywhere near the rate that the college wage pre-
mium has increased over the equivalent periods, if they have risen at all.
Finally, the education mismatch literature cited here shows that the wage
premium from a college degree comes mainly from getting access to a job
that requires college-level skills. A college graduate in a job that requires
only high school skills earns little more than a high school graduate in that
job. That should remind us of the fallacy of composition. For an individual
17When we compare recent graduates to all graduates, we see changes in the college premium over
time. In 1995, for example, the median bachelor-degree graduates ages 24 to 34 earned 30% more than
the median high school graduates in the same age group, but all bachelor-degree graduates over age 25
earned 50% more than all high school grads over age 25. In 2011, the equivalent gures were 50% and
65%, respectively.
271SKILL GAPS, SHORTAGES, AND MISMATCHES
to secure a college degree may make perfect sense because doing so will pay
off. Whether, in the absence of any evidence of an increased demand for
college-level skills, society as a whole should send a higher percentage of
high school students on to college expecting that they will all earn that same
premium is not obvious.
Student Achievement
Comparative evidence on student achievement gets arguably the most atten-
tion in assertions that skill gaps exist that are caused by the U.S. education
system. Here the argument is that U.S. students are not learning as much as
those elsewhere. New data sets, especially the Trends in Math and Science
Study (TIMSS), which began in 1995, and the Programme for International
Student Assessment (PISA), which began in 2000, showed that the United
States is about in the middle of the rankings of other countries on the vari-
ous measures of student achievement.
Loveless (2012) noted that many observers interpret this evidence as
showing that U.S. students score worse than many other countries with
higher rankings, but he noted that no statistically signicant difference exist
between the U.S. scores and those for countries often several positions
higher in the rankings. He also noted that, despite claims to the contrary,
the U.S. position has not been declining over time relative to other coun-
tries. In the rst of the international comparison tests in 1964, for example,
the United States scored 11th out of 12 countries, but U.S. student perfor-
mance on subsequent tests has being going up in absolute terms in the both
the TIMSS and the PISA tests (Loveless 2011). Scores in other countries
have been going up as well, however, so the U.S. position relative to other
countries has remained about the same—near the middle.
The fact that Asian countries such as Singapore and Korea and the Chi-
nese cities of Hong Kong and Shanghai have now risen to the top ranks in
these studies has received considerable attention, but what is not clear is
how much credit for those high scores should go to their schools. As many
as 70% of students in those countries also attend after-hours tutoring classes,
known as juku in Japan, cram schools in Korea, and the “shadow education
system” generally (Bray and Lykins 2012). Surely this supplemental set of
practices affects student achievement.
The newest and most powerful evidence on skills across countries comes
from the Organisation for Economic Co-Operation and Development
(OECD) Programme for International Assessment of Adult Competencies
(PIAAC). This is an international comparison of worker skills—literacy,
numeracy, and problem solving in an IT context—rather than student skills.
It is assessed directly using tests of representative, random samples of the
workforce from each country. On this assessment, the United States ranked
17th in literacy, 22nd in numeracy, and 14th in problem solving out of the
24 countries participating, far below average. A related assessment of the
U.S. position comes from a recent study of the wage premiums associated
272 ILR REVIEW
with the PIAAC skills data. It found that the highest skill premiums were in
the United States, which is consistent with a situation in which these skills
are in relatively short supply compared to other countries (Hanushek, Schw-
erdt, Wiederhold, and Woessmann 2013).
Nevertheless, academic preparation in school—which happened decades
before for the average respondent—was not closely related to the poor
showing of U.S. workers relative to workers in other countries, despite asser-
tions in the popular press. The PIAAC authors noted that the United States
has a more-educated workforce than average and, more important, that the
relative position of the United States in student achievement is much higher
than its position in worker skills. As previously noted, U.S. students come
out about average in most international comparisons of academic achieve-
ment and also did so in earlier decades, when most of the PIAAC respon-
dents were in school. (Many of the worst-performing countries in this study
are not in the PIAAC study, however, which keeps the two from being strictly
comparable.) Together, these results suggest that something is happening
to the reading and numeracy skills of U.S. workers after they leave school
that is different from what is happening in other countries.
More puzzling in these data are the differences in scores by age group.
Overall, younger workers perform much better on the PIAAC skills tests
than do their older counterparts, suggesting the widespread upgrading of
skills over the generations—except in the United States. Younger U.S. work-
ers score about the same as older workers: a little worse than those ages 25
to 44, a little better than those older than 44 years. Here again, the relative
academic achievement of U.S. school leavers compared to other countries
has remained about the same over time, so the explanation cannot be
declining academic achievement of U.S. school leavers over time. The best
explanation for these results points to events after workers leave school,
such as more upskilling on the job (workplace training or education jobs
that keep such skills sharp) now for younger workers in other countries
compared to the United States, although some part of the gap may have to
do with immigration practices—immigrants to the United States have had
lower cognitive skills than immigrants to other countries (Kahn 2004).
The PIAAC also asked respondents about the hiring criteria associated
with qualications (academic degrees and similar credentials) in their cur-
rent job and compared them to the qualications of the respondent. Hiring
criteria are not identical to job requirements, of course, and we would expect
the former to rise when applicants are plentiful, as they have been in recent
years. Nevertheless, the results in Figure 1 show more than 50% more work-
ers in OECD countries believed that their current job could be performed
with fewer qualications (education and credentials) than are currently
required by their employer than workers who reported that more qualica-
tions are actually required. The U.S. scores are similar to the OECD average.
A different but related complaint is the idea that the United States is declin-
ing in the number of graduates with college degrees it produces relative to
other countries. As with the assertions cited earlier that the United States is
facing some labor force shortage, these complaints appear to misread the
273SKILL GAPS, SHORTAGES, AND MISMATCHES
most basic evidence. The absolute number of graduates is certainly not declin-
ing; as previously noted, the number of bachelor degrees increased from just
under 1.3 million in 2000 to 2001 to just over 1.7 million in 2010 to 2011, or a
39% rise, and the number of associate degrees increased by even more, from
580,000 in 2000 to 2001 to more than one million in 2011 to 2012, or a 62%
increase, with a population growth of only 11% (NCES 2014). True, the
United States does not lead the world in having the highest percentage of col-
lege graduates in the population, but it has not done so for some time. Russia,
Canada, Japan, and Israel have a higher percentage of graduates. (Strictly
speaking, the measure used for comparison is for postsecondary degrees, not
all of which are comparable to bachelor degrees; OECD 2013: Chart A1.1.)
What is more distinctive is that the United States is one of the few coun-
tries where the younger cohort, ages 25 to 34, does not have more educa-
tion than the older cohort. Israel, Finland, Germany, and Brazil are also in
that situation. The shortfall in the U.S. cohort is attributable to men because
women in this age group do have more degrees than does the older cohort.
One complication with the conclusion that the younger cohort in the
United States is behind those in other countries is that U.S. citizens are far
more likely to earn college degrees later in life than are those in other
0510 15 20 25 30 35
Russian Federaon
Netherlands
Italy
Austria
Estonia
Denmark
Spain
Norway
Slovak Republic
Canada
Ireland
OECD Average
Germany
Japan
Flanders (Belgium)
Poland
Sweden
Finland
France
Czech Republic
United States
Korea
England/N. Ireland (UK)
Higher level needed
Lower level sufficient
Figure 1. Hiring Requirements Compared to Performance Requirements
Source: OECD Survey of Adult Skill; Konstantinos Pouliakas performed the analysis on behalf of the
European Centre for the Development of Vocational Training. Used with permission from Konstantinos
Pouliakas.
Notes: Respondent answers to the question, “Thinking about whether this qualication is necessary for doing
your job satisfactorily, which of the following statements would be most true?” and where “this qualication”
is dened by a preceding question: “Still talking about your current job: If applying today, what would be the
usual qualications, if any, that someone would need to GET this type of job?” Country averages were
obtained using the piaactab command in STATA; Public Use les (PUFs) of OECD PIAAC data used.
274 ILR REVIEW
countries, so the U.S. position in the rankings should change.18 Whether or
how the U.S. position matters in rankings such as these is not obvious, how-
ever. We return to this question below.
Where the United States does not look good in a comparative sense is in
our graduation rate from college, for which we are second from the bottom,
with roughly 60% of those who enroll graduating (OECD 2013: Chart A4.1).
Again, this low position may reect the fact that so many U.S. students are
part-time (58%) and may have eventually graduated after the data were col-
lected. We do lead the world in spending per degree, however (ibid.: Chart
B1.2). Individuals in the United States bear a much greater share of the
costs of postsecondary education than average, but individuals in the United
Kingdom, Korea, and Chile pay an even higher share (ibid.).
Summarizing and Assessing Evidence of Skill Problems
The contemporary reports arguing that skill problems exist in the workforce
offer little in the way of compelling evidence to support that claim, and the
evidence that they present is often contradictory. Objective evidence from
government data and from other sources does not support any of the claims
about skill problems. In fact, the evidence appears to be compelling that the
United States is experiencing exactly the opposite problem—a substantial
skill mismatch in the form of individuals with more education than their cur-
rent jobs require and a surplus of educated and skilled workers who cannot
nd jobs at all, let alone jobs appropriate for their education and skill level.
Employer-led complaints about U.S. skill problems, broadly dened, are
not new. Alchian, Arrow, and Capron (1958) responded to assertions about
perceived shortages of engineers and scientists in the 1950s, noting, among
other things, that those complaining had a poor understanding of how
labor markets actually work.19 More recently, Magnum (1990) reported on
manufacturing employers who complained since World War II that they
faced a skill shortage and chronicled a history of assertions that skill prob-
lems were going to occur, all of which turned out to be false. Rosenthal
(1982) assessed complaints from manufacturers about a skill shortage com-
ing from the 1981 recession and found no evidence of it. His recommenda-
tions—that employers seeing difculty in hiring should raise their wages,
train more to develop talent internally, or substitute technology for labor to
reduce their demand—are standard textbook solutions that are largely
absent from the current discussions.
Reports of employer complaints have accelerated dramatically in recent
years, however, especially since the Great Recession with its unprecedented
18For example, 13% of full-time U.S. undergraduate students in four-year colleges are over age 25;
http://nces.ed.gov/programs/coe/indicator_csb.asp, Figure 1.
19The framework they used, still useful today, considered many possible denitions of shortage, includ-
ing nontechnical use. They described the idea of a “dynamic shortage,” in which employers cannot hire
more candidates at the wage they are currently paying, as among the most common situations causing
complaints. In such cases, time lags and morale complications can ensue while the employer adjusts to
higher wages, but Alchian, Arrow, and Capron did not see public policy interventions as helpful.
275SKILL GAPS, SHORTAGES, AND MISMATCHES
and paradoxical (for the complainers) surfeit of available talent. Indeed,
much of the attention that complaints about skill problems have received
during the recession in the popular press may well have been because of the
“man-bites-dog” aspect of the claims: With all these people looking for jobs,
why can’t these employers ll theirs?
What these complaints from employers about skill problems might actu-
ally mean remains elusive because the employer-driven studies are so poorly
designed and conducted that discerning what real information they contain
is difcult. A safe conclusion does seem to be, however, that, to the extent
they are real, employer complaints probably reect concerns about hiring
experienced candidates because so many more of them are in the job mar-
ket than school leavers; because concerns about school leavers per se focus
on maturity, not academic skills; and because many—perhaps most—
employer problems with hiring appear to be self-inicted (such as offering
inadequate pay and training).
Survey questions that ask, “Are you having difculty nding the candi-
dates you need to ll your vacancies?” are maddeningly ambiguous. Simply
following the many steps and issues that should be considered in textbook
descriptions of recruiting and the selection processes could constitute dif-
cult. We also know that self-serving biases should inate complaints—assert-
ing that the candidates have something wrong with them is much easier
than acknowledging that our own practices are at fault.
One explanation for the large number of employer-based complaints
may just be that they are part of the broader rise in recent years in business
lobbying to inuence public opinion and policy in ways that benet employ-
ers (Waterhouse 2014). The employer complaints in the IT sphere, for
example, go hand in hand with lobbying efforts to increase access to foreign
workers with H1B status. The basic conicts between labor and manage-
ment are not far from the surface in many of these exchanges, with employer
groups arguing for more supply and, in turn, lower wages and greater
employer power in the employment relationship, while labor and employee
groups (albeit considerably fewer of these) argue for the opposite. The
reports on skill problems from consulting rms are also consistent with a
self-serving explanation: Their business models are based on helping their
business clients address perceived problems. Highlighting or even asserting
problems and then offering solutions to them are common practices. The
rm HRMarketer.com (n.d.), which provides advice on how to sell consult-
ing services to human resource departments, offers a related explanation. It
recommends that vendors produce white papers and other reports address-
ing big questions to build credibility with clients, and one of those big ques-
tions they note is the shortfall of talent.
A second explanation is that the rise in employer complaints about dif-
culties in hiring reects something real, even if it is not caused by any change
in the supply of skill in the labor market. Hiring may well be more difcult
now simply because employers have to do much more hiring these days
because of widespread and substantial declines in employee tenure (see
Hollister 2011 for a survey), which translates into more frequent vacancies
276 ILR REVIEW
and more hiring to ll them. The decline of lifetime employment and the
associated rise of lateral hiring have been underway for some time, especially
in larger organizations (Cappelli 1999). The fact that the decline in tenure is
disproportionately associated with larger rms where promotion from within
had been more common (Bidwell 2013) may have an even bigger effect on
hiring to the extent that it undermines promotion-from-within systems or is
a marker for their decline; when employees who have been promoted from
within leave unexpectedly, lling their jobs from within may be difcult
because no internal candidates may be ready for advancement.
A decline in promotion-from-within systems also increases hiring chal-
lenges substantially by expanding the range of jobs through which hiring
takes place from standard entry-level jobs, which require minimal skills, to
virtually every position in the organization. The much wider range of hiring
would, in turn, require nding experienced candidates who either are or
have worked elsewhere. Finding such candidates with the precise skills
needed to ll these jobs is much more challenging than nding inexperi-
enced school leavers for entry-level jobs. Systematic data on external hires
are difcult to obtain, but a proprietary survey from the same pool of employ-
ers over time (Career Xroads 2013) found that 72% of their positions were
lled from outside in 2007. Outside hiring dropped sharply during the reces-
sion to less than half of all placements but then rose back up to 58% in 2012.
Hiring directly from college—a proxy for entry-level jobs—accounted for
only 5.5% of all new employees hired by these rms. Presumably most of the
other 94.5% were experienced hires. Another proprietary survey of large
employers (12,000 average employees) reported that only 15% of the posi-
tions they lled in April 2014 could have been performed by an applicant
just out of high school or college.20 The common assumption that employer
complaints about skill problems necessarily refer to jobs lled by new gradu-
ates is hard to square with these data.
When employers do hire from college, the evidence suggests that aca-
demic skills are not their primary concern. A survey of employers conducted
by the Chronicle of Education (2013) showed that work experience was the
crucial attribute that employers wanted, even for students who had yet to
work full-time, and that the relevance of coursework to the job in question
was just not that important (see Figure 2).
Few employer reports of skill problems ask what exactly employers are
looking for in the candidates who they cannot nd, but the evidence in Fig-
ure 2 suggests that it is work experience, even for school leavers. Just as the
European employers (described previously) explained about why they did
not train, these employers are seeking to get the skills they need through
hiring, and a signicant part of those skills appears to come from work
experience. Evidence for the idea that work experience is the issue for
20Survey conducted by Recruiting Trends magazine, 2014. No data are available about how many recent
graduates, if any, were hired into those positions. The average position lled had 74 applicants, and the
average hire had nine years of experience. A total of 227 employers responded to the Recruiting Trends
magazine survey. Thanks to the editors for making the results available.
277SKILL GAPS, SHORTAGES, AND MISMATCHES
employers comes from recent surveys of employers that were conducted by
the Minnesota Department of Labor for occupations thought to be difcult
to ll because of skill shortfalls. Even for engineering and IT jobs, employ-
ers reported no difculty nding qualied college graduates. “Rather, appli-
cants lacked hyper-specialized experience or a unique blend of skills that
could be extremely difcult to nd even when plenty of people apply”
(Leibert 2013: 8). Overall across the occupations covered in the survey,
employers reported that “skills mismatches were sometimes found to be
related to inadequate levels of experience, but not to inadequate levels of
education” (Leibert 2013: 6). Hiring candidates who already have skills, in
turn, becomes more important once employers no longer have training and
apprenticeship programs as an alternative.
Credible evidence on employer-provided training in the United States is
remarkably hard to come by, especially recently. The data we do have suggest
that in 1979 young workers received on average about 2.5 weeks of training
per year. By 1991, U.S. Census data found only 17% of all employees reported
that they received any formal training that year. Several surveys of employers
around 1995 indicated that somewhere between 42% and 90% of employers
offered some training (the lower percentage indicating more programmatic
training), with the amount of training an individual received per year averag-
ing just under 11 hours. The most common training topic was workplace
safety.21 The gures also include whatever training vendors provide when
they bring in new equipment: “Here’s how to work this copier.”
Figure 2. Employer Preferences for Applicant Attributes
Source: Chronicle of Higher Education (2012).
21For a review of the data on the extent of training, see Lynch and Black (1998) and Frazis, Gittleman,
and Joyce (2000).
278 ILR REVIEW
These data are now almost 20 years old, and government sources have
provided little new data. In 2011, Accenture surveyed U.S. employees and
found that only 21% had received any employer-provided formal training in
the previous ve years. To be clear, that means almost 80% had received no
training in ve years, and no doubt many of those had received no training
in the years before that, either.
The most important source for craft-based skills has been apprenticeship
programs. Data on the extent of U.S. apprentices and programs are not exten-
sive, but the Department of Labor data on apprentice programs registered
with it and that meet certain quality standards show a sharp decline from 2002
to 2012—from roughly 33,000 programs to roughly 21,000—and an even
steeper decline in the number of apprentices—from roughly 500,000 in 2003
to approximately 280,000 in 2012.22 The approximately 50,000 annual gradu-
ates of these programs is a drop in the bucket of a labor force of 160 million.
Whether these measures are representative of trends in all apprenticeship
programs is not clear because nonregistered apprenticeship programs cer-
tainly exist and the data for those other programs are not available in any sys-
tematic fashion. The Department of Labor has pushed since 2004 to identify
and register more programs (OECD 2012: 4), so if the rate of registration has
been increasing, then the decline in actual apprentice programs would be
even greater than the data here suggest. Further, the quality of the appren-
ticeship programs is not necessarily constant. Bilginsoy (2003) noted that, in
the construction industry, union–management joint apprenticeship programs
have been in decline as the unions running them have declined. They are
being replaced by employer-based programs that do not perform as well as
the joint apprenticeship programs, perhaps because employers are in more of
a hurry to get the trainees into the jobs they need to ll now and many more
apprentices leave the employer-led programs before they complete them.
One area in which employer complaints about shortfalls in the supply of
skill have unique credibility is in the skilled trades. Especially for smaller
employers, an important source of workers who had at least basic trade skills
was vocational education programs in high schools. Figure 3 illustrates
course-taking among students who graduated from public school. Begin-
ning in 1990, the number of vocational courses taken declined precipitously,
especially compared to the rise in other subject areas.
Within vocational education curricula, industrial arts, which includes
skilled trades and other mechanical skills, declined even faster. The average
number of credits taken per student in that subject area fell to half from 2000
to 2005.23 The United States already had the least proportion of vocational
22Curiously, the number of “program completers” remained unchanged over this period, at just over
50,000, even though the number of programs and the number of apprentices declined sharply. That
translates to about two completers per apprentice program each year (see ETA 2012).
23Here the NCES (2014c: Table H125) data suggest slightly different trends depending on the occu-
pational classications. The percentage of students earning credits in the construction area fell only
slightly from 1990 to 2009, by 0.7%. The percentage in manufacturing, which includes machinist skills,
fell by 9.5%. Interestingly, business coursework declined by almost 20%, but at the undergraduate level
it expanded considerably during the same period. These data say only whether students took any courses,
in contrast to the BLS data, which indicate how many courses they took.
279SKILL GAPS, SHORTAGES, AND MISMATCHES
education in secondary school education of any of the industrialized coun-
tries although the fact that U.S. data is not available makes comparisons dif-
cult (OECD 2013: Table A1.5). Employers who relied on vocational education
to provide their new hires would, indeed, nd a shrinking pool of candidates.
This appears to be the best evidence that something has changed in the sup-
ply of labor to manufacturing employers.
Different countries have made different choices about who is responsible
for aspects of skills (see Biavaschi 2012 for an overview). The United States
is already the outlier in having fewer vocational education options and in
generally requiring individuals to get and pay for a larger amount of job
skills themselves after leaving high school. Much has been made of the
importance of community colleges as the providers of skills in the United
States, and the Obama administration has followed previous administra-
tions in calling for an increase in their role (White House n.d.). Community
colleges are funded largely at the local level, however, and whether they will
Figure 3. Changes in High School Credits by Subject Area
Source: NCES (2009).
Note: CTE, Career and Technical Education.
280 ILR REVIEW
have the resources to take on the training that employers do not want to do
is an open question.24
A nal explanation for employer complaints, related to the previous dis-
cussion, is the highly specic and idiosyncratic nature of contemporary hir-
ing requirements. We might assume, for example, that a job opening for a
position such as a machinist would be reasonably standard, but that turns
out not to be the case.25 This situation is consistent with Lazear’s (2003)
skill-weights view, in which the appearance of rm-specic skills can be gen-
erated by a unique mix of general skills in job requirements and different
priorities attached to those skills across employers.
If job requirements across employers are highly specic and somewhat
unique, then the supply of available workers is constrained, perhaps sharply
so. This creates the possibility of monopsony for employers, an upward-slop-
ing supply curves for labor. Monopsonists who pay equivalent workers the
same wage—nondiscriminating monopsonists—nd wage levels rising
quickly when they try to hire more workers, and they end up hiring fewer
workers and producing less than if labor supply were more elastic (see Man-
ning 2003). Job applicants might nd it very difcult to determine which
skills they should acquire when requirements are specic and varied. In the
absence of very good information about jobs and mobility, they may nd
themselves skilled but also unemployed, or at least underemployed.
An obvious solution to the monopsony problem, and indeed to virtually
all the skill problems reported by employers, is to increase training and pro-
duce the skilled workers they want themselves. Anecdotally, employers often
express the view that they cannot afford to train employees for fear that they
will then be hired away at a higher wage by competitors, the textbook prob-
lem of providing general skills. Whether that is simply a rationalization or
an opinion based on evidence is important to know. The counter to that
view is the extensive literature, starting in the recent period with Acemoglu
and Pischke (1998), on how the provision of general skills can be structured
to address that problem.
Conclusion
Overall, the available evidence does not support the idea that serious skill
gaps or skill shortages exist in the U.S. labor force. The prevailing situation in
24Tracing course-taking per se at community colleges is difcult, and we know that a great many stu-
dents who go there to pick up job skills do not intend to pursue a degree. Although most associate
degrees from community colleges are indeed in very applied elds, one-third are in liberal arts (NCES
2013: Table 312), a far greater proportion than in bachelor-degree programs. The explanation appears
to be that community colleges also play a role as feeders into four-year degree programs and liberal arts
majors represent that pathway.
25To get a simple sense of the job requirements for machinist positions, I investigated the vacancies
posted on the Monster.com website for May 13, 2014. Machinist openings are subdivided into 13 catego-
ries based on differences in job requirements. Within the “CNC” subcategory, every job required prior
experience (one also required further training through a community college certicate program). Most
required experience with particular brands of CNC machine tools. Every job had at least one require-
ment that did not appear in any of the other job openings.
281SKILL GAPS, SHORTAGES, AND MISMATCHES
the U.S. labor market, as in most developed economies, continues to be skill
mismatches in which the average worker and job candidate have more educa-
tion than their current job requires. Persistently high levels of unemployment
reect the fact that job seekers still outnumber the available job openings.
Although a bigger supply of cheaper labor would certainly be useful to
employers, it is not clear that such a situation would be useful for the country
as a whole, and any claims to that effect should be examined carefully.
What the apparent rise in employer complaints about skill problems rep-
resents is not clear, in part because of the poor quality of information pre-
sented as part of the complaints. No doubt some component of the
complaints is simply an effort to secure policy changes that will lower labor
costs. Some component of the complaints may well represent real problems
associated with changes in employer behavior, such as greater outside hiring
(and associated increases in employee turnover) and reduced training and
internal development. Some of these changes might be driven by the behav-
ior of other employers. For example, increases in turnover may be driven by
the hiring practices of other employers, and smaller employers who in the
past were able to meet their skill needs by hiring skilled apprentices away
from larger employers may nd that no one is available to hire now that
those apprenticeship programs are gone. Efforts to hire skills rather than to
build them from within would create much more specic and variable job
requirements across employers, which would vastly increase not only the dif-
culty in hiring but also the experience of having to raise wages above cur-
rent levels to nd appropriate candidates.
The implications that follow from these conclusions are important to
consider. The proposals in the various employer-led reports that we need to
increase student academic achievement, reduce drop-out rates, and gener-
ally improve the quality of secondary school education are difcult to object
to, but these goals are already accepted. The challenge has been how to
achieve them. The interest in having applicants with more accountability,
motivation, and maturity is also hard to argue with, but these attributes are
also typically associated with growing up. How schools can accelerate that
process and make 18-year-olds act like 28-year-olds is not clear.
A more important implication of many of the reports considered here is
the underlying shift they represent in responsibility for the provision of a
workforce. The notion that employers are responsible for recruiting and
retaining employees and then training and developing them to meet chang-
ing skill needs—textbook human resource concepts—is hard to see in any
of these reports, even when the results show that employer practices are
driving much of their perceived skill problems.
The view that emerges from these arguments is that the responsibility for
developing the skills that employers want has been transferred from the
employer to the job seekers and schools. Such a transfer of responsibility
would be profound in its implications. Schools, at least as traditionally envi-
sioned, are not suited to organizing work experience, the key attribute that
employers want. Nor are they necessarily good at teaching work-based skills.
282 ILR REVIEW
Those skills are easiest and cheapest to learn in the workplace through
apprentice-like arrangements that can be found not only in the skilled
trades but also in elds such as accounting and medicine. Unlike in the
classroom, in the workplace problems to practice on do not have to be cre-
ated. They exist already, and solving them creates value for others. Observa-
tion and practice are also easiest to do where productive work is being done,
and employment creates incentives and motivation that typical classrooms
cannot duplicate.
At the postsecondary level, where job skills have traditionally been offered
in U.S. schools, the shift in the responsibility for developing skills from employ-
ers to students has important implications for the students and their families.
The obvious implication is that, because the costs must be paid for up-front,
individuals without the capital to pay the fees lose access to those skills, again
in contrast to earlier periods when employers provided more opportunities for
on-the-job learning and training. That shift in responsibility also pushes the
risk onto students and their families. The employers who are calling for more
STEM graduates, for example, are not offering to hire the students who are
now starting such programs when they graduate. The students and their fami-
lies who pursue specic education in hopes of obtaining jobs at the end are
taking a substantial nancial risk should those jobs not be there.
Proposals such as those in the state of Florida would push students toward
vocational majors by shifting state funds to college majors from which
employers say they want to hire (typically STEM elds) and away from
majors from which they do not (Anderson 2011). Governments do not have
a particularly good record of forecasting where jobs will be years in advance,
however, and students and their families will bear the costs when those fore-
casts are wrong.
More generally, if the labor market is not enticing students to pursue par-
ticular elds, should public policy push them to do so? Manufacturers, for
example, have long complained about the shortage of students interested in
machinist training programs and have asserted that the cause is that schools
and guidance counsellors do not advocate for those programs. But the pay
for such jobs has declined by 20% in real terms over the past two decades
while the skill requirements for those jobs have shifted toward computer use,
a eld with better pay. Moreover, the number of machinist jobs has already
declined by 20% during that period (the total number of jobs in the econ-
omy has increased by 40%) and is expected to decline further (Cappelli
2012). The reasons for the decline in the number of students taking voca-
tional education courses that could prepare them for manufacturing jobs
merits further attention, but we should not assume that it is independent
from the attractiveness of the jobs offered at the end of those programs.
Assuming that the government cannot become the stafng agency for
employers, are we faced with a future in which employers are frustrated
because they cannot nd the specic skills they want to hire at the same
time that jobs seekers and especially school leavers cannot get the skills that
employers really want because no one will give them initial work
283SKILL GAPS, SHORTAGES, AND MISMATCHES
experience? Some employers may see a market opportunity for getting skills
more cheaply by training even if competitors want to hire away the workers
they train. We know that employers can provide even general skills training
if the arrangements are structured appropriately (Acemoglu and Pischke
1998), as we see in temporary help agencies (Autor 2001), with tuition reim-
bursement plans (Cappelli 2004), and in most professional service rms
(consulting, accounting, etc.) that turn liberal arts graduates into consul-
tants and bankers, for example, through on-the-job training. Traditional
apprenticeship programs operate in the same way.
Assuming that not enough employers gure out how to manage that
option, are alternative arrangements possible? The arrangements favored
by the school-to-work movement in the 1990s may still have merit. In such
programs, the boundary between school and work blurred. Employers
helped schools to provide work-based learning content that supplemented
academic material and offered learning opportunities in the workplace that
were not necessarily paying jobs. The employer’s incentive to participate
was the ability to identify promising students to hire before the students
ever went on the job market, and the costs of the programs were minimal.
The topic of skill problems in the United States represents something
reasonably new for labor market and public policy discussions. It is difcult
to think of another labor market issue for which academic research or even
research using standard academic techniques has played such a small role,
where parties with a material interest in the outcomes have so dominated
the discussion, where the quality of evidence and discussion has been so
poor, and where the stakes are potentially so large. The perspectives and
interests of employees and students have been almost completely absent
from these discussions. Little testing of the assumptions behind arguments
has been done, and the costs and benets of various proposals have not
been considered. This situation seems to be unique to the United States.26
One factor in the relative lack of academic research on these topics no
doubt has been the lack of information and data about skills per se. The
standard classication of job requirements into “knowledge, skills, and abil-
ities” reminds us that education, which has served as a proxy for skills in
most discussions, maps onto only part of the “knowledge” category, leaving
the other attributes of job requirements out of the picture. Many important
reasons exist for being concerned about education, but seeing it as the
equivalent of skill is certainly a mistake. One of the unfortunate conse-
quences of using education as the proxy for skill is that schools, the provid-
ers of education, are then seen as the mechanism for dealing with skill
problems, leaving on-the-job training and experience out of the story.
26For example, the Global Agenda Council (GAC) of the World Economic Forum on Employment,
whose members are predominately outside the United States, considered the notion that skill gaps and
skill shortages might exist as irrelevant. Indeed, an employer representative from outside the United
States asserted that “no one makes such arguments.” See Global Agenda Council on Employment (2014)
for a global view.
284 ILR REVIEW
A nal lesson for the U.S. academic research community from the cur-
rent discussion of skill problems is that, in the absence of clear, objective
research ndings, advocates for the existence of skills problems can easily
make claims that are simply assertions and that even a casual acquaintance
with real evidence would indicate are false. Perhaps the many organizations
and foundations that have supported the advocacy-oriented approaches
reviewed here might be persuaded to help change the debate by supporting
real research on the role of skills in the U.S. economy.
References
Accenture. 2012. Solving the skills paradox: Seven ways to solve your critical skills gap.
Accessed at http://www.accenture.com/SiteCollectionDocuments/PDF/Accenture-
Solving-the-Skills-Paradox.pdf.
Acemoglu, Daron. 2002. Technical change, inequality, and the labor market. Journal of Eco-
nomic Literature 40(1): 7–72.
Acemoglu, Daron, and David Autor. 2011. Skills, tasks, and technologies: Implications for
employment and earnings. In David Card and Orley Ashenfelter (Eds.), Handbook of
Labor Economics, Vol. 4a, pp. 1043–71. Amsterdam: North Holland.
———. 2012. What does human capital do? A review of Goldin and Katz’s The Race Between
Education and Technology. Journal of Economic Literature 50(2): 426–63.
Acemoglu, Daron, and Jorn-Steffen Pischke. 1998. Why do rms train? Theory and evidence.
Quarterly Journal of Economics 113(1): 79–119.
Acemoglu, Daron, and Fabrizio Zilibotti. 2001. Productivity differences. Quarterly Journal of
Economics 116(2): 563–606.
Acemoglu, Daron, David Autor, David Dorn, Gordon H. Hanson, and Brendan Price. 2014.
Return of the Solow paradox? IT, productivity, and employment in U.S. manufacturing.
NBER Working Paper No. 19837. Cambridge, MA: National Bureau of Economic Research.
Alchian, Armen A., Kenneth Arrow, and William M. Capron. 1958. An economic analysis of
the market for scientists and engineers. Rand Corporation. Accessed at http://www.rand
.org/content/dam/rand/pubs/research_memoranda/2005/RM2190.pdf.
Allen, Jim, and Rolf van der Velden. 2001. Educational mismatches versus skill mismatches:
Effects on wages, job satisfaction, and on-the-job search. Oxford Economic Papers 53(3):
434–52.
Anderson, Zac. 2011. Rick Scott wants to shift university funding away from some degrees.
Herald-Tribune, October 10. Accessed at http://politics.heraldtribune.com/2011/10/10/
rick-scott-wants-to-shift-university-funding-away-from-some-majors/.
Aspen Institute. 2003. Grow faster together or grow slowly apart. Domestic Strategy Group.
Accessed at http://www.aspenwsi.org/wordpress/wp-content/uploads/GrowFast.pdf.
[ASTD] American Society for Training and Development. 2012. Bridging the skills gap: Help
wanted, skills lacking. Why the mismatch in today’s economy? Accessed at http://nist
.gov/mep/upload/Bridging-the-Skills-Gap_2012.pdf.
Atkinson, Richard C. 1990. Supply and demand for scientists and engineers: A national crisis
in the making. Science 248(4954): 425–32.
Atkinson, Robert D., and Luke A. Stewart. 2013. The real story on guestworkers in the high-
skill U.S. labor market. Washington, DC: Information Technology and Innovation Foun-
dation. Accessed at http://www2.itif.org/2013-guestworkers-high-skill-labor-market.pdf.
Autor, David H. 2001. Why do temporary help rms provide free general skills training? Quar-
terly Journal of Economics 116: 1409–48.
———. 2010. The polarization of job opportunities in the U.S. labor market: Implications for
employment and earnings. Washington, DC: Center for American Progress and the Ham-
ilton Project.
Autor, David H., Lawrence F. Katz, and Melissa S. Kearney. 2008. Trends in US wage inequal-
ity: Revising the revisionists. Review of Economics and Statistics 90(2): 300–323.
285SKILL GAPS, SHORTAGES, AND MISMATCHES
Autor, David H., Lawrence F. Katz, and Alan B. Krueger. 1998. Computing inequality: Have
computers changed the labor market? Quarterly Journal of Economics 113(4): 1169–213.
Autor, David H., Frank Levy, and Richard J. Murnane. 2003. The skill content of recent tech-
nological change: An empirical exploration. Quarterly Journal of Economics 118(4): 1279–
333.
Badillo Amador, Lourdes, Ángel López Nicolás, and Luis E. Vila. 2012. The consequences on
job satisfaction of job–worker educational and skill mismatches in the Spanish labour
market: A panel analysis. Applied Economics Letters 19(4): 319–24.
Baert, Stijn, Bart Cockx, and Dieter Verhaest. 2013. Overeducation at the start of the career:
Stepping stone or trap? Labour Economics 25: 123–40.
Barnichon, Regis, Michael Elsby, Bart Hobijn, and Ays¸egül S¸ahin. 2012. Which industries are
shifting the Beveridge curve? Monthly Labor Review. Accessed at http://www.bls.gov/
opub/mlr/2012/06/art2full.pdf.
Barnow, Burt S., John Trutko, and Jaclyn Schede Piatak. 2013. Occupational Labor Shortages:
Concepts, Causes, Consequences, and Cures. Kalamazoo, MI: Upjohn Institute.
Bayer Corporation. 2013. Is there a STEM workforce shortage? October 22. Accessed at
http://www.bayerus.com/News%5CNewsDetail.aspx?ID=DCF6FA59-C9B5-2F14-
A7AB69927FB04B0B.
Beaudry, Paul, David A. Green, and Benjamin M. Sand 2013. The great reversal in the demand
for skill and cognitive tasks. NBER Working Paper No. 18901. Cambridge, MA: National
Bureau of Economic Research.
Bennett, Jessica, and Seamus McGuinness. 2009. Assessing the impact of skill shortages on
the productivity performance of high-tech rms in Northern Ireland. Applied Economics
41(6): 727–37.
Berg, Ivar. 1970. Education and Jobs: The Great Training Robbery. Westport, CT: Praeger.
Biavaschi, Costanza, Werner Eichhorst, Corrado Giulietti, Michael J. Kendzia, Alexander
Muravyev, Janneke Pieters, Nuría Rodríguez-Planas, Ricarda Schmidl, and Klaus F. Zim-
mermann. 2012. Youth unemployment and vocational training. IZA Discussion Paper No.
6890. Bonn, Germany: Institut zur Zukunft der Arbeit.
Bidwell, Matthew J. 2013. What happened to long-term employment? The role of worker
power and environmental turbulence in explaining declines in worker tenure. Organiza-
tion Science 24(4): 1061–82.
Bilginsoy, Cihan. 2003. The hazards of training: Attrition and retention in construction
industry apprenticeship programs. Industrial and Labor Relations Review 57(1): 54–67.
Bluestone, Barry, and Mark Melnik. 2010. After the recovery: Help needed. Civic Ventures.
Accessed at http://www.encore.org/les/research/JobsBluestonePaper3-5-10.pdf.
Borjas, George. 2010. Labor Economics. New York: McGraw-Hill.
Boston Consulting Group. 2013. The U.S. skills gap: Could it threaten the U.S. manufactur-
ing renaissance? Accessed at https://www.bcgperspectives.com/content/articles/lean_
manufacturing_us_skills_gap_could_threaten_manufacturing_renaissance/.
Braverman, Harry. 1974. Labor and Monopoly Capital: The Degradation of Work in the
Twentieth Century. New York: NYU Press.
Bray, Mark, and Chad Lykins. 2012. Shadow Education: Private Supplementary Tutoring and
Its Implications for Policy Makers in Asia. Mandaluyong City, Philippines: Asian Develop-
ment Bank.
Brencˇicˇ, Vera. 2010. Do employers respond to the costs of continued search? Oxford Bulletin
of Economics and Statistics 72(2): 221–45.
British Chambers of Commerce. 2011. Skills for business: More to learn? Accessed at http://
www.britishchambers.org.uk/assets/downloads/policy_reports_2011/Skills%20for%20
business%20-%20more%20to%20learn.pdf.
Brynjolfsson, Erik, and Andrew McAfee. 2011. Race against the Machine. Lexington, MA: Digi-
tal Frontier Press.
[BLS] Bureau of Labor Statistics. 2013. Labor force projections to 2022: The labor force par-
ticipation rate continues to fall. Monthly Labor Review. Accessed at http://www.bls.gov/
opub/mlr/2013/article/labor-force-projections-to-2022-the-labor-force-participation-
rate-continues-to-fall-1.htm.
286 ILR REVIEW
———. 2014a. Job openings and labor turnover survey. Accessed at http://data.bls.gov/
timeseries/JTU30000000JOL?data_tool=XGtable.
———. 2014b. Job openings and labor turnover survey highlights, March 2014. Chart 1.
Number of unemployed persons per job opening. Accessed at http://www.bls.gov/web/
jolts/jlt_labstatgraphs.pdf.
———. 2014c. Occupational outlook handbook. Accessed at http://www.bls.gov/ooh/most-
new-jobs.htm.
Cachon, Gerard, and Christian Terwiesch. 2006. Matching Supply with Demand. New York:
McGraw-Hill.
Cainarca, Gian Carlo, and Francesca Sgobbi. 2012. The return to education and skills in Italy.
International Journal of Manpower 33(2): 187–205.
Cappelli, Peter. 1995. Rethinking the “Skills Gap.” California Management Review 37(4): 108–
24.
———. 1999. The New Deal at Work: Managing the Market-Driven Workplace. Boston: Har-
vard Business School Press.
———. 2001. Employer participation in school-to-work activities and involvement with
schools. The National Center for Postsecondary Improvement, Stanford University.
———. 2003 Will there really be a labor shortage? Organizational Dynamic 32(3): 221–33.
———. 2004. Why do employers pay for college? Journal of Econometrics 121(1–2): 213–41.
———. 2012. That pesky skill shortage in manufacturing. HR Executive. Accessed at http://
www.hreonline.com/HRE/view/story.jhtml?id=534354686.
Career Advisory Board. 2011. Research report: Career advisory board job preparedness indi-
cator. DeVry University, and Harris Interactive Interviewing. Accessed at http://careerad
visoryboard.com/public/uploads/2011/11/Job-Preparedness-Indicator-Research
-Report.pdf.
Career Xroads. 2013. Sources of hire 2013: Perception is reality. Accessed at http://www
.careerxroads.com/news/SourcesOfHire2013.pdf.
Carnevale, Anthony. 2005. The coming labor and skills shortage. Training & Development
59(1): 37–41.
Carnevale, Anthony P., Nicole Smith, and Jeff Strohl. 2010. Help wanted: Projections of jobs
and education requirements through 2018. Washington, DC: Center on Education and the
Workforce, Georgetown University. Accessed at http://cew.georgetown.edu/jobs2018/.
Cascio, Wayne F. 2008. To prosper, organizational psychology should bridge application and
scholarship. Journal of Organizational Behavior 29(4): 455–68.
Challenger, John A. 2003. The coming labor shortage. Futurist 37(5): 24–28.
Chambers, Elizabeth G., Mark Foulon, Helen Handeld-Jones, Steven M. Hankin, and
Edward G. Michaels. 1998. The war for talent. McKinsey Quarterly (3): 44–57.
Charette, Robert M. 2013. The STEM crisis is a myth. IEEE Spectrum, August. Accessed at
http://spectrum.ieee.org/at-work/education/the-stem-crisis-is-a-myth (August 2014).
Chronicle of Higher Education. 2012. The role of higher education in career development:
Employer perceptions. Survey. Accessed at https://chronicle.com/items/biz/pdf/
EmployersSurvey.pdf.
Computer and Technology Industry Association. 2012. State of the IT skills gap. Accessed at
http://www.wired.com/wiredenterprise/wp-content/uploads//2012/03/Report_-_
CompTIA_IT_Skills_Gap_study_-_Full_Report.sb_.pdf.
[CBI] Confederation of British Industry. 2013. Tomorrow’s growth report. Accessed at
http://www.cbi.org.uk/campaigns/skills-for-growth/tomorrows-growth-report/.
[CBO] Congressional Budget Ofce. 2011. CBO’s labor force projections through 2021.
Accessed at http://www.cbo.gov/sites/default/les/cboles/ftpdocs/120xx/doc12052
/03-22-laborforceprojections.pdf.
[CVTS] Continuing Vocational Training Statistics. 2013. European Commission, Brussels.
Accessed at http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Continu
ing_vocational_training_statistics.
Davis, Steven J., R. Jason Faberman, and John C. Haltiwanger. 2012. Recruiting intensity dur-
ing and after the Great Recession: National and industry evidence. American Economic
Review 102(3): 584–88.
287SKILL GAPS, SHORTAGES, AND MISMATCHES
———. 2013. The establishment-level behavior of vacancies and hiring. Quarterly Journal of
Economics 128(2): 581–622.
Deloitte and the Manufacturing Institute. 2011. Boiling point? The skills gap in U.S. manu-
facturing. Accessed at http://www.themanufacturinginstitute.org/~/media/A07730B2A
798437D98501E798C2E13AA.ashx.
Duncan, Greg, and Saul Hoffman. 1981. The incidence and wage effects of overeducation.
Economics of Education Review 1(1): 75–86.
Dychtwald, Ken, Tamara J. Erickson, and Robert Morrison. 2006. Workforce Crisis: How to Beat
the Coming Shortage of Skills and Talent. Boston: Harvard Business School Press.
Edwards, Richard C. 1979. Contested Terrain: The Transformation of the Workplace in the
Twentieth Century. New York: Basic Books.
Engineering Employer’s Federation 2013. Skills for Growth. Accessed at http://www.eef.org
.uk/representation/%20campaigns/route-to-growth/default.htm.
[ETA] Employment and Training Administration. 2012. Department of Labor Performance
Plan FY 2011. Washington, DC: U.S. Department of Labor.
Federation of Small Businesses. Raising the standards: An FSB skills survey. Accessed at
http://www.fsb.org.uk/policy/assets/skills%20report%20nal.pdf.
Feldstead, Alan, Duncan Gallie, Francis Green, and Hande Inanc. 2013. Work Intensication in
Britain: First Findings from the Skills and Employment Survey 2012. London: Centre for Learn-
ing and Life Chances in Knowledge Economies and Societies.
Frazis, Harley, Maury Gittleman, and Mary Joyce. 2000. Correlates of training: An analysis
using both employer and employee characteristics. Industrial and Labor Relations Review
53(3): 443–62.
Freeman, Richard B. 1976. The Overeducated American. New York: Academic Press.
———. 2006. Is a great labor shortage coming? Replacement demand in the global economy.
NBER Working Paper No. 12541. Cambridge, MA: National Bureau of Economic Research.
Global Agenda Council on Employment. 2014. Matching skills and labour market needs:
Building social partnerships for better skills and better jobs. World Economic Forum,
Geneva, Switzerland.
Goldin, Claudia, and Lawrence F. Katz. 2008. The Race between Education and Technology. Cam-
bridge, MA: Harvard University Press.
Gottschalk, Peter, and Morton Hansen. 2003. Is the proportion of college workers in non-
college jobs increasing? Journal of Labor Economics 21: 449–71.
Government Accounting Ofce. 1998. Assessment of the Commerce Department’s report on
worker demand and supply. T-HEHS-98-144. Washington, DC.
Hanushek, Eric A., Guido Schwerdt, Simon Wiederhold, and Ludger Woessmann. 2013.
Returns to skills around the world: Evidence from PIAAC. NBER Working Paper No.
19762. Cambridge, MA: National Bureau of Economic Research.
Harrington, Paul E., and Andrew M. Sum. 2010. College labor shortages in 2018? New Eng-
land Journal of Higher Education, November 8. Accessed at http://www.nebhe.org/thejour
nal/college-labor-shortages-in-2018/.
———. Forthcoming. Recent projections of labor shortages through 2018: From great reces-
sion to labor shortages? A critical look at the evidence. Monthly Labor Review.
Hollister, Matissa. 2011. Employment stability in the US labor market: Rhetoric versus reality.
Annual Review of Sociology 37: 305–24.
HRMarketer.com. No date. Marketing and selling in the human resource marketplace: Win-
ning strategies and tactics. Accessed at http://www.hrmarketer.com/selling_to_HR.pdf.
Institute of Directors. 2010. Reforming the skills system: Lessons learned the hard way.
Accessed at http://www.iod.com/mainwebsite/resources/document/reforming_the%
20_skills_system_lessons_learned_the_hard_way_jul11.pdf.
Jacobs, Jerry A., and Linda Sax. 2014. STEM majors, the liberal arts and the great recession. Paper
presented at the American Educational Research Association Annual Meeting, April 5.
Jacoby, Sanford. 1983. Employing Bureaucracy: Managers, Unions, and the Transformation
of Work in the 20th Century. New York: Columbia University Press.
Jobs Council. 2012a. Prepare the American workforce to compete in the global economy.
Accessed at http://www.jobs-council.com/recommendations/prepare-the-american-
workforce-to-compete-in-the-global-economy/.
288 ILR REVIEW
———. 2012b. Ensure that Americans have the right skills for today. Accessed at http://www
.jobs-council.com/recommendations/ensure-that-americans-have-the-right-skills-for-
today%E2%80%99s-jobs/.
Joyce, Mary, and David Neumark. 2001. School-to-work programs: Information from two sur-
veys. Monthly Labor Review 124: 38–50.
Kahn, Lawrence M. 2004. Immigration, skills, and the labor market: International evidence.
Journal of Population Economics 17(3): 501–34.
Kampelmann, Stephan, and François Rycx. 2012. The impact of educational mismatch on rm
productivity: Evidence from linked panel data. Economics of Education Review 6: 918–31.
Katz, Lawrence F., and Kevin M. Murphy. 1992. Changes in relative wages, 1963–1987: Supply
and demand factors. Quarterly Journal of Economics 107(1): 35–78.
Keefe, Jeffrey H. 1990. Numerically controlled machine tools and worker skills. Industrial and
Labor Relations Review 44(3): 503–19.
Krueger, Alan B. 1993. How computers have changed the wage structure: Evidence from
micro data. Quarterly Journal of Economics 108: 33–60.
Lazear, Edward P. 2003. Firm-specic human capital: A skills-weighted approach. NBER
Working Paper No. 9679. Cambridge, MA: National Bureau of Economic Research.
Lazear, Edward P., and James R. Spletzer. 2012. The United States labor market: Status quo or
a new normal? NBER Working Paper No. 18386. Cambridge, MA: National Bureau of
Economic Research.
Leibert, Alessia. 2013. Are skilled workers scarce? Evidence from employer surveys in Min-
nesota. Minnesota Department of Employment and Economic Development. Accessed at
http://mn.gov/deed/images/SecondRoundReportSkillsGap.pdf.
Levine, Marc V. 2013. The skills gap and unemployment in Wisconsin: Separating fact from
ction. University of Wisconsin at Milwaukee, Center for Economic Development.
Liu, Yujia, and David B. Grusky. 2013. The payoff to skill in the third industrial revolution.
American Journal of Sociology 118: 1330–74.
Loveless, Tom. 2011. 2010 Brown Center Report on American Education: How Well Are American
Students Learning? Vol. II, no. 5. Brown Center on Educational Policy. Washington, DC:
The Brookings Institute.
———. 2012. 2011 Brown Center Report on American Education: How Well Are American Students
Learning? Vol. III, no. 1. Brown Center on Educational Policy. Washington, DC: The
Brookings Institute.
Lynch, Lisa M., and Sandra E. Black. 1998. Beyond the incidence of employer-provided train-
ing. Industrial and Labor Relations Review 52(1): 64–81.
Magnum, Stephen. 1990. Impending skill shortages: Where is the crisis? Challenge 33(5): 46–52.
Manning, Alan. 2003. Monopsony in Motion: Imperfect Competition in Labor Markets. Princeton,
NJ: Princeton University Press.
Manpower. 2013. The talent shortage survey. Accessed at http://www.manpowergroup.us/
campaigns/talent-shortage-2012/pdf/2012_Talent_Shortage_Survey_Results_US_
FINALFINAL.pdf.
Mavromaras, Kostas, and Seamus McGuinness 2007. Education and skill mismatches in the
labour market: Editor’s introduction. Australian Economic Review 40(3): 279–85.
Mavromaras, Kostas, Seamus McGuinness, Nigel O’Leary, Peter Sloane, and Zhang Wei.
2013. Job mismatches and labour market outcomes: Panel evidence on university gradu-
ates. Economic Record 89(286): 382–95.
McGuinness, Seamus, and Peter J. Sloane. 2011. Labour market mismatch among UK gradu-
ates: An analysis using REFLEX data. Economics of Education Review 30: 130–45.
McKinsey Global Institute. 2012. The world at work: Jobs, pay and skills for 3.5 billion people.
Accessed at http://www.mckinsey.com/insights/employment_and_growth/the_world_
at_work.
Mishel, Lawrence, Heidi Shierholz, and John Schmitt. 2013. Don’t blame the robots: Assess-
ing the job polarization explanation of growing wage inequality. November 9. Washing-
ton, DC: Economic Policy Institute.
Mortensen, Dale T. 1986. Job search and labor market analysis. Handbook of Labor Economics 2:
849–919.
289SKILL GAPS, SHORTAGES, AND MISMATCHES
Mosca, Irene, and Robert E. Wright. 2013. Is graduate under-employment persistent? Evi-
dence from the United Kingdom. IZA Discussion Paper No. 6177. Bonn, Germany: Insti-
tut zur Zukunft der Arbeit.
National Academies of Science, Engineering, and the Institute of Medicine. 2007. Rising
above the gathering storm: Energizing and employing America for a brighter economic
future. Accessed at http://www.nsf.gov/attachments/105652/public/NAS-Gathering-
Storm-11463.pdf.
[NCES] National Center for Education Statistics. 2009. Career/technical education (CTE)
statistics. Figure 1. Accessed at http://nces.ed.gov/surveys/ctes/gures/g_01.asp.
———. 2013. Condition of education: Undergraduate degree elds. Washington, DC: NCES.
———. 2014a. Table 318: Degrees conferred by degree-granting institutions, by control of
institution and level of degree: 1969–70 through 2010–11. Washington, DC: NCES.
———. 2014b. Table 438: Median annual earnings of full-time year-round workers 25 years
old and over, by highest level of educational attainment and sex: 1990 through 2011.
Washington, DC: NCES.
———. 2014c. Table h125: Average number of credits and percentage of total credits that
public high school graduates earned during high school, by curricular area: 1990, 2000,
2005, and 2009. Career/Technical Education Statistics. Washington, DC: NCES.
National Center on Education and the Economy. 1990. America’s choice: High skills or low
wages! The report of the Commission on the Skills of the American Workforce. National
Center on Education and the Economy, Rochester, NY.
National Commission on Excellence in Education. 1983. A nation at risk: The imperative for
educational reform. April. Accessed at http://datacenter.spps.org/uploads/SOTW_A_
Nation_at_Risk_1983.pdf.
[NRC] National Research Council. 2001. Building a Workforce for the Information Econ-
omy. Washington, DC: National Academies Press.
———. 2008. Future skill demands: A workshop summary. Washington, DC: National Acad-
emies Press.
Neumark, David, Hans Johnson, and Marisol Cuellar Mejia. 2011. Future skills shortage in
the U.S. economy? NBER Working Paper No. 17213. Cambridge, MA: National Bureau of
Economic Research.
Ng, Ying Chu. 2003. Earnings and “skill” allocation in the Canadian labor market. Interna-
tional Journal of Manpower 24(8): 964–80.
[OECD] Organisation for Economic Co-operation and Development. 2010. Skills beyond
school. Accessed at http://www.oecd.org.
———. 2012. OECD note on quality apprenticeships for the G20 Task Force on employment.
Accessed at http://www.oecd.org/els/emp/OECD%20Apprenticeship%20Note%20
26%20Sept.pdf.
———. 2013. Education at a glance 2013: OECD indicators. Paris: OECD Publishing.
Accessed at http://dx.doi.org/10.1787/eag-2013-en.
Osterman, Paul, and Andrew Weaver. 2014. Skills and skill gaps in manufacturing. In Richard
Locke and Rachel Wellhausen (Eds.), Production in the Innovation Economy, pp. 17–50.
Cambridge, MA: MIT Press.
Pouliakas, Konstantinos. 2014. Matching skills and labour market needs in Europe: Challeng-
ing the usual suspects. Address given at the Programme for the International Assessment
of Adult Competencies Launch Event for the PIAAC Survey in Greece, Mapping Skills
Shortages, Planning the Future, March 31, Athens.
President’s Council of Advisors on Science and Technology. 2012. Prepare and inspire: K–12
education in science, technology, engineering and math (STEM) education for Ameri-
ca’s future. Accessed at http://www.whitehouse.gov/sites/default/les/microsites/
ostp/pcast-engage-to-excel-nal_2-25-12.pdf.
President’s Council on Jobs and Competitiveness. 2012. Roadmap to renewal. Accessed at
http://www.jobs-council.com/recommendations/summary-of-road-map-to-renewal-
report/.
PricewaterhouseCoopers (PwC). 2012. Facing the talent challenge: Global CEO survey.
Accessed at http://www.pwc.com/gx/en/ceosurvey/2012/key-ndings/hr-talent-strate
gies.jhtml.
290 ILR REVIEW
Rogerson, Richard, Robert Shimer, and Randall Wright. 2005. Search-theoretic models of the
labor market: A survey. Journal of Economic Literature 43: 959–88.
Rosenthal, Neal H. 1982. Shortages of machinists: An evaluation of the information. Monthly
Labor Review 105(7): 31–36.
Rothstein, Jesse. 2012. The labor market four years into the crisis: Assessing structural expla-
nations. ILR Review 65(3): 467–500.
Salzman, Hal, Daniel Kuehn, and B. Lindsay Lowell. 2013. Guestworkers in the high-skill US
labor market. EPI Brieng Paper No. 359. Washington, DC: Economic Policy Institute.
[SCANS] Secretary’s Commission on Achieving Necessary Skills. 1991. What work requires of
schools. U.S. Department of Labor. Accessed at http://wdr.doleta.gov/SCANS/what-
work/whatwork.pdf.
[SHRM] Society for Human Resource Managers. 2003. SHRM survey results. Alexandria, VA:
SHRM. June 22. Available at http://www.shrm.org/research/quarterly/captureIDs.
asp?type=0304aging_essay.asp.
Sicherman, Nachum, and Oded Galor. 1990. A theory of career mobility. Journal of Political
Economy 98(1): 169–92.
Smith, James P., and Finis Welch. 1978. The overeducated American? A review article. Techni-
cal report. Santa Monica, CA: Rand Corporation.
Spleen, Thomas Luke. 2013. Recent college graduates in the US labor force: Data from the
Current Population Survey. Monthly Labor Report 136(2): 3–13.
Stull, William J., and Nicholas M. Sanders (Eds.). 2003. The School-to-Work Movement: Origins
and Destinations. New York: Praeger.
Sutherland, John. 2012. Qualications mismatch and skills mismatch. Education and Training
54(7): 619–32.
Tinbergen, Jan. 1974. Substitution of graduate by other labor. Kyklos 27: 217–26.
U.S. Chamber of Commerce. 2006. The state of American business 2006. Washington, DC.
U.S. Department of Commerce. 1997. America’s new decit: The shortage of information
technology workers. Washington, DC: Ofce of Technology Policy.
Vaisey, Stephen. 2006. Education and its discontents: Overqualication in America, 1972–
2002. Social Forces 85(2): 835–64.
Venneberg, Donald L., and Barbara W. Eversole. 2010. The Boomer Retirement Time Bomb: How
Companies Can Avoid the Fallout from the Coming Skills Shortage. Santa Barbara, CA: Praeger.
Walsh, William D. 1977. The short-run behavior of skilled wage differentials. Industrial and
Labor Relations Review 30(3): 302–13.
Waterhouse, Benjamin C. 2014. Lobbying America: The politics of lobbying from Nixon to
NAFTA. Princeton, NJ: Princeton University Press.
White House. n.d. Building American skills through community colleges. Accessed at http://
www.whitehouse.gov/issues/education/higher-education/building-american-skills-
through-community-colleges.
Wolf, Alison, Liam Aspin, Edmund Waite, and Katerina Ananiadou. 2010. The rise and fall of
workplace basic skills programmes: Lessons for policy and practice. Oxford Review of Edu-
cation 36(4): 385–405.
Zicklin, Gilbert. 1987. Numerical control machining and the issue of deskilling: An empirical
view. Work and Occupations 14(3): 452–66.
CopyrightofIndustrial&LaborRelationsReviewisthepropertyofCornellUniversityand
itscontentmaynotbecopiedoremailedtomultiplesitesorpostedtoalistservwithoutthe
copyrightholder'sexpresswrittenpermission.However,usersmayprint,download,oremail
articlesforindividualuse.
