stage. These could include community-based nonprofits, associations of employers, and
workforce development organizations, among others.
Of course, demand projections often have some degree of error, especially since labor
demand can shift in directions that are not easily predictable from recent trends.
Therefore, state plans should also indicate the extent to which the education and training
provided is general and likely portable across specific sectors if such unanticipated
demand shifts occur. The best plans will also include funding and/or technical assistance
for employers who might need modest retraining either for newly hired or incumbent
workers who don’t exactly fit their current skill needs.
Thus, state plans should provide
both for occupation- and industry-specific training but also for mechanisms that generate
flexible responses to unanticipated demand shifts.
Broader Measures to Support Employment-Based Training
The grants would be used to encourage more responsiveness to the labor market at
community or four-year colleges. For instance, the grants could be used to expand high-
quality CTE programs in high school, career counseling at colleges, and also to
encourage educational institutions to expand instructional capacity in high-demand areas,
based on labor market data, where such capacity is often now lacking. Indeed, states
could be rewarded for tying their subsidies for community colleges to rates of certificate
or degree completion, especially in sectors of strong demand. The integration of
developmental or remedial education with occupational training could be encouraged,
along with other proven efforts to reduce dropout rates.
Some funds would be available to pay for tax credits or technical assistance to good-
paying employers participating in sectoral training programs and other efforts to upgrade
their incumbent workers; a model for this technical assistance might be the
Manufacturing Extension Partnership (MEP) program that now helps manufacturers
upgrade workplace performance and productivity. More broadly, states should indicate
that their education and workforce systems are also part of broader economic
development plans to assist industry development and employment growth, especially in
geographic areas that are currently underserved (McGahey and Vey, 2009; Bartik, 2010).
Funding Direct Services for Trainees
Grants to states would then pay for some direct service provision that is not already
available to Pell grantees and other lower- or middle-income postsecondary students.
These services could include tuition payments for coursework leading to certification in
the relevant fields, by both prospective and incumbent employees, who are not eligible
For instance, the explanatory power of BLS projections relative to subsequent labor market trends has
been challenged by Carnevale et al. (2010), who show that their own projections have stronger predictive
power over time. Alternatively, some analysts (e.g., Bishop and Carter, 1991) claim that the BLS
projections are systematically biased and tend to underpredict the growth in labor market demand for
education over time.
For instance, in Uchitelle’s example of welders cited above, many employers do not claim that they
cannot find welders in general, but seek a particular type of welding training which might not be widely
available at any point in time.
C# HTML5 PDF Viewer SDK deployment on IIS in .NET
and set the “Physical path” to the place where you store XDoc.PDF.HTML5 Viewer Demo. Pool Defaults…" in the right panel, and set the value "Enable 32-Bit pdf thumbnail generator; pdf thumbnail viewer
for Pell grants;
stipends for paid work experience under apprenticeships, internships,
and other forms of college work study in these fields; and supportive services, such as
child care for low-income parents. Small federal programs that already provide such
funding, such as the Child Care Access Means Parents in School (CCAMPIS) program,
or the Job Location and Development Program (JLDP) that provides paid work to
students off-campus under the Federal Work Study program, could be effectively
expanded and perhaps even incorporated into such efforts.
Promoting Sustainability through Leveraging of other Existing Funding Sources
States would only receive grant money if they provide better services to students and
better incentives to institutions as part of lasting systemic plans to improve the better
matching of less-educated or disadvantaged workers with good jobs over time.
To encourage more lasting plans, states would have to generate plans to sustain their
efforts over time, using other public and private sources of funds.
The new program should leverage other recent and current funding efforts, especially if
the states can indicate how they are building on the progress generated from those other
efforts. For instance, besides TAACCCT, the proposed fund could complement activities
funded by the U.S. Department of Labor through recent competitive grant programs such
as the High Growth and Emerging Industries Job Training Initiative and the Workforce
Innovations for Regional Economic Development (WIRED) grants to regions. It could
also complement the efforts of several national foundations, such as the National Fund
for Workforce Solutions; and others aimed at community colleges and/or states to
improve degree completion rates as well as career pathways to local labor markets, such
as “Achieving the Dream,” “Shifting Gears” and “Breaking Through.”
It would build
on activities already begun in many states (NGA Center for Best Practices, op. cit.) to
more closely link their education and workforce activities (including those funded by
WIA) to economic development, and also on major new workforce initiatives like the No
Worker Left Behind program recently implemented in Michigan. That program provided
training funds to dislocated workers who were being trained in community colleges for
jobs in industries where high future growth is expected.
Most importantly, the grants would hopefully encourage much better use of the enormous
sums of federal money recently invested in the Pell grant program by the Obama
Administration, and of very large state subsidies to public colleges as well, by raising
certificate or degree completion rates among grant recipients that are well-matched to
good jobs in the labor market. As such, this program would not reinvent the wheel or
duplicate other efforts, but would build on them. The grants would encourage states to
Low-income students may not be eligible for Pell grants if they get training from a provider that is not an
accredited 2-year or 4-year college. Other reasons for disqualification include felony drug convictions.
The National Fund has been established by the Annie E. Casey, Ford, Hitachi and other foundations. It
now includes 300 funders for sectoral training projects in 24 communities. The “Achieving the dream,”
“Shifting Gears” and “Breaking Through” programs have been funded by the Gates, Lumina, Joyce and
other foundations . None of these efforts have been rigorously evaluated to date.
combine currently disparate and uncoordinated funding efforts into more effective
education and workforce systems, better matched to state and local labor market demand.
Private funding sources should be leveraged as well. Indeed, since employers would
benefit to some extent from these programs, they should be willing to contribute some
modest funding, perhaps through their industry associations or through dedicated funds
from state payroll taxes.
Implemented in this fashion, the program would hopefully generate the kinds of lasting
systemic changes at the state level that have apparently been induced by other federal
grant programs recently, like the Race to the Top fund in K-12 education or the
expansions of Unemployment Insurance (UI) eligibility under the Unemployment
Insurance Modernization Act (UIMA) provisions in the recent federal stimulus bill.
Evidence Base and Evaluation
The criteria provided above are in part based on the evidence about what creates a
successful training program, but the state plans should explicitly indicate the extent to
which their proposals reflect evidence of cost-effectiveness based on rigorous research
analysis, such as the best studies cited above.
The capacity to conduct rigorous evaluations of their own programs at both the
institutional and state levels would be required as a condition of receipt of funding.
Where specific programs are being set up or expanded, experimental evaluations based
on randomized controlled trials would be considered most appropriate. Alternatively,
states could also generate nonexperimental evaluations using appropriate methods, either
for specific programs and policies or for their overall efforts more broadly.
of grant applicants to conduct evaluations should be verified by the contractor selected by
the Departments of Labor and Education to conduct the evaluation. Renewal of these
grants would at least partly depend on the extent to which evaluation evidence indicates
success in expanding employment opportunities and earnings for the targeted groups.
IV. Expected Costs and Benefits
What kinds of broader impacts might we expect from the kind of policy initiative
described above? The potential impacts are extremely hard to gauge, since our program
would provide direct payments for limited amounts of new services while also heavily
leveraging others that are already being made in very large amounts (like Pell grants and
current state subsidies to community college students) to render them more effective.
Any such estimates are quite speculative about the numbers served directly (by receiving
services funded under the grants) or indirectly (by receiving services already funded that
These taxes have been used to fund incumbent worker training programs in a variety of states
For instance, states could propose to use difference-in-differences (DD) or regression discontinuity
designs (RDD) across institutions or regions that have implemented different kinds of specific policies and
practices. Analysis at the state level could be performed, for instance, using interrupted time series
are now more effectively delivered) and how many would be positively affected by the
Subject to these caveats, I have calculated the likely costs of the program and the benefits
to the nation in higher earnings associated with this proposal, using two different
approaches and some different assumptions within each. Table 3 presents the estimates
associated with the first approach, which I regard as the stronger of the two (in terms of
the evidence based for the calculations provided). Details of all of the calculations used to
generate these estimates appear in the technical appendix to this paper.
The table presents estimates of benefits from the program under different assumptions for
rates of program completion and “fade-out” (which is the rate at which earning gains
from the program decay over time). The estimates use the average cost of the training
programs from the Sectoral Employment Impact Study described previously, as well as
estimated earnings gains generated there.
The first set of cost and benefit estimates focuses on the training services provided
directly to workers. I assume that $1.5B of the grant money is spent per year on direct
services at an average cost of $6,000 per trainee, implying that up to 250,000 individuals
might be directly served by this program in any given year. Thus, if the program lasts five
years, as many as 1.25 million individuals could cumulatively receive services over time.
In the case where gains fade out at 5 percent per year and 50 percent of participants
complete the program (column 1), the present value of total benefits for a year of such
program would be 5.75 billion dollars, implying a benefit-to-cost ratio of 3.8. Even with
the conservative assumptions of 20 percent fade-out and 50 percent completion, the
benefits of the program still exceed the cost. Of course, since these calculations assume
no effects (either on costs or benefits) of other expenditures that might generate other
services and reforms in how community colleges and local workforce systems operate,
they are likely lower bounds to the true impacts of the program. Since we presume that
changes in community colleges and other parts of the education and workforce systems in
states would be much broader in scope, those served both directly and indirectly could
number much higher than this.
Alternatively, I also have made some calculations (described in the appendix) that
assume we could increase the rate of credential attainment by 10 percentage points either
among all community college students or Pell grant recipients, enabling them to at least
earn labor market certificates that generate the kinds of impacts that were estimated in
SEIS (but, on average, less than associates degrees). I also assume that a competitive
program of this magnitude could award grants to 10 states of average size, covering up to
one-fifth of the nation’s postsecondary population, especially in community colleges.
A program of this magnitude, if successfully implemented over a number of years, could
generate positive impacts in earnings that are very cost-beneficial. The estimates of total
and net benefits from assuming higher rates of certification attainment among community
college or Pell recipients respectively, are larger than the estimates that use the cost of
providing sectoral training directly. Using the baseline assumption of 5 percent fade-out
and 50 percent completion, the present value of total benefits associated with a 10
percentage point increase in certificate attainment would be over $11B in the former and
over $8B in the latter case; the benefit-to-cost ratios would be 5.7 and 4.1 respectively.
It is important to note that the benefits associated with these programs would likely
accrue not only to private individuals and their employers, but to the public sectors at all
levels (federal, state and local) as well. For instance, if the earnings of disadvantaged and
dislocated workers could be raised, it would likely raise income tax revenues at these
levels; and it would reduce poverty-related public expenditures over time associated with
the high unemployment, high crime rates and poor health of these populations (Belfield
and Levin, 2007; Holzer et al., 2007; Holzer, 2010a).
Specifically, successful efforts to
raise employment and earnings for these populations should reduce currently high
expenditures in Medicaid, Unemployment Insurance and the funding of criminal justice
systems for these populations. By raising tax revenues and reducing public expenditures
in these areas, the new federal expenditures for these grants would be at least partly
offset, and state budget deficits might be reduced.
Given the high job vacancy as well as unemployment rates that currently exist, we would
expect these efforts to reduce unemployment rates by enabling more workers to be
matched both to jobs that are now being created and to new ones. For instance, one can
imagine that perhaps one-tenth of a percentage point of the unemployment rate could be
eliminated by the higher credentials achieved by students and workers under this program
per year, and up to one-half of a percentage point after five years.
If at least some grants
are awarded within the first year, and because many of the certificates supported can be
earned with training of fairly short duration, impacts should begin to be observed within
two years of the program start date.
Thus, at least the potential exists for some quite small but effective expenditure of funds
to have major impacts on the employment and earnings of the nation’s disadvantaged
workers, as well as on the productivity outcomes that underlie them.
V. Questions and Concerns
One of the first questions that might be asked about the proposal described above is the
extent to which it overlaps with or duplicates efforts funded now by the Workforce
Investment Act (WIA). This legislation now funds state and regional workforce
investment boards (or WIBs), which in turn fund employment services at local One-Stop
offices as well as limited amounts of job training. The WIBs currently engage in some
state and local planning, and sometimes cooperate with community colleges and other
educational institutions in meeting local labor demand (Besharov and Cottingham, 2011).
While there is some overlap between what now exists and what we propose, some key
differences exist as well. For instance, the new grants would explicitly call for plans to be
This calculation assumes that up to 150,000 of those earnings credentials because of the new program
can fill job slots that are vacant at any point in time.
built around targeting underserved populations for jobs in key economic sectors. They
would fund many more services than are now generally allowable or available under
They would more actively and directly engage state and local higher education
institutions, and would incent these and other institutions to be more responsive to trends
in labor demand than they are now.
But it would also be important that any new grants programs not be used to reduce
formula funding right now for WIA. Given the extent to which WIA funds have already
been drastically cut over the past years and decades, and how tight those resources are for
the cost-effective local employment services and training that they now fund, it is
important that these new grants constitute a net new addition of resources, and not further
cannibalize some important existing programs.
Another question involves the extent to which the Departments of Education and Labor
can jointly implement a grants program, at the federal level as well as the state and local
levels. Some precedent exists for such efforts, such as the administration of funds to local
areas under the School to Work Opportunities Act (STWOA) of the mid-to-late 1990s,
and many more recent examples of cooperation between the two agencies (as well as the
Department of Health and Human Services). But the grants program would also create
new opportunities for local “silos” to be opened up and more comprehensive systems to
be built. Indeed, the new grants would create incentives for this to occur, and states
would be awarded grants at least partly on the basis of the extent to which such systems
are built. Renewal of the grants would also be an opportunity to judge which states have
generated effective partnerships between workforce and education agencies on the
As usual, one of the concerns about such a proposal would be the extent to which
successful smaller efforts in the past can be replicated and scaled up nationally. While
this concern is valid, the fact that grants will be awarded competitively, and that there are
strong models to be replicated that have themselves already achieved some significant
scale, give us somewhat greater confidence on this issue.
Another concern is whether or not the more specific occupational or sectoral skills in
which prospective workers are being trained will generate long-term labor market
rewards for them, especially after they leave the job in question. More broadly, one might
also wonder whether the occupational training provided would be too narrow – in other
words, should we invest significant resources in specific sectoral training, given the
likelihood of unexpected shifts over time in the composition of labor demand in a very
technologically dynamic world, or only more general training?
See Holzer (2009) for evidence that WIA has been cut by nearly 90 percent in real terms since
1980,while the scope of its services has expanded and while the workforce has grown by roughly half.
O’Leary et al. (2004) report that funding for employment and training services in the U.S., as a percent of
GDP, lags behind that of almost every other industrial country. On the other hand, some uses of the
workforce innovation funds now being allocated in WIA might now be subsumed under the new grants
While this concern is certainly valid, the extent of the potential problem can be limited in
a number of ways. First, all workers at community colleges should get some mix of more
general skills as well as those that are more narrowly tailored to a particular occupation
or sector; to the extent that workers move across jobs and sectors over time, at least some
of the skills would thus be transferable, depending on how far from the original sector
their mobility takes them. The same is true if labor demand shifts away from sectors
where such demand is now strong and where such training is readily provided.
Second, a more effective education and workforce system should itself lead to more
effective employer adaptations to such shifts in demands for skills. If, for instance,
employers need welders, but mostly of a different type than they currently find among
prospective workers (as Louis Uchitelle of the New York Times has recently suggested), a
more effective workforce system should make it easier for employers to modestly retool
their incumbent or prospective workers, and this would limit the difficulties associated
with specific skills training.
Another concern is whether the current fiscal environment will allow for even the kinds
of modest new expenditures that have been proposed above. With proposals for large cuts
in federal discretionary nondefense spending, and in particular for job training, now being
generated, it might not be a very auspicious time to propose some increases. On the other
hand, recent evidence suggests that expenditures in education are not quite as vulnerable
to cuts at the federal level; and those tied to job creation and employer needs might be
less vulnerable to cuts, if they enjoy some bipartisan support (especially from major
employers and industry associations).
It might be possible to reallocate some of these funds from other employment and
One possible source of funding for new competitive grants is revenues
from H-1B visa fees. H-1B visas are visas for high-skill workers. The revenues from
these visas are intended to be used for training American workers.
If alternate funding is not available, the cost of the program might be scaled back initially
and ramped up more slowly as successes become more apparent and political support
grows over time.
It is also important that assistance be targeted primarily on students and workers with the
greatest need – in other words, disadvantaged youth and adults (who are capable of being
effectively trained and can handle more technical material when necessary) as well as
dislocated workers. Economic development efforts at the state level might be used to
provide public assistance to employers or middle-class workers who want a “free ride”
but who could afford to pay for the relevant education and training themselves. While
these efforts may not reach the hardest-to-employ populations (such as those with the
poorest numeracy and literacy skills and other barriers to work), they should be judged at
least partly on their targeting of groups in need, as well as the other criteria listed above.
A recent report by the U.S. Government Accountability Office (USGAO, 2011) indicates several dozen
small federal employment and training programs that overlap, to some extent, with WIA and might produce
significant savings if carefully consolidated.
Finally, we need to note the overall weakness of the US job market, both in the short-
term and the longer-term. Insufficient aggregate demand and uncertainty seem to be
limiting overall job creation and our recovery from the Great Recession, while new
technologies and global forces might do so over the longer term as well (Blinder, 2006;
Freeman, 2007; McKinsey Global Institute, 2011). This proposal is not designed to
address a broader set of problems that seem to be deterring employers from creating large
numbers of jobs, as they did in the 1980s and 1990s.
On the other hand, the need for enhancements in worker skills and the quality of jobs
created remains, and perhaps becomes even stronger, in a tepid labor market. And the
ability of these markets to absorb workers with higher skill levels and higher pay over the
longer term should not be doubted, even when aggregate employment outcomes are
To raise employment levels and earnings in the US, I propose a new set of grants to fund
more effective education and workforce systems at the state level, which would
especially be more supportive of firms that create good-paying jobs, hopefully
encouraging them to create even more. The grants would fund partnerships of employers,
training providers and intermediaries at the state and local levels, as well as a range of
specific services and activities. Criteria have been laid out for the awarding of grants,
including the extent to which they target underserved populations and growing sectors,
the extent of services provided, the extent to which other sources of public and private
funding are leveraged, and plans for rigorous evaluation of outcomes and impacts.
The proposal builds on a body of existing research that indicates the success and potential
for further targeting training towards firms and sectors that create good-paying jobs. The
proposal does not reinvent the wheel or duplicate existing programs, but is specifically
designed to build on efforts that are already underway in many places.
I believe that, if effectively designed and implemented, such a grants program could
significantly improve the employment rates as well as earnings of targeted groups over
the next few years and beyond as well.
Documents you may be interested
Documents you may be interested