What works in workforce development—and how can it work better?


For the past two years, a bipartisan group of researchers and analysts convened by the American Enterprise Institute, the Brookings Institution, and the Harvard Kennedy School’s Project on Workforce has reviewed the evidence on the effectiveness of the United States’ federal-state workforce education and training system. Our group—the Workforce Futures Initiative (WFI)—has reached some surprising and hopefully useful conclusions about how our nation can improve its investments in job training.

The good news is that federal spending on workforce development—including the Workforce Innovation and Opportunity Act (WIOA) system—improves disadvantaged worker outcomes. The bad news is that improvements are quite modest. In the words of one of our group members, we appear to be “stuck in a low-resource, low-efficacy” equilibrium. Small benefits at low levels of funding discourage higher levels of investment; yet without additional funding, it is unlikely we’ll see substantial improvement. At the same time, students and workers lack other options to finance training—for instance, Pell grants do not cover noncredit or shorter-term training efforts. The WIOA system, and the workers who use it, are caught in a policy catch-22.

Greater public investment in workforce development programs is needed. But these additional investments should be targeted toward programs and practices that have proven successful and can be scaled, or that provide information that is critical to diagnose the needs of a rapidly changing labor market. Examples include sectoral employment programs, job counseling and supportive services, improvements to data systems to better track program performance and improve our understanding of changing skill demands, and pilot programs to test ways of increasing system flexibility and innovation.

Sectoral employment programs substantially improve employment and wage outcomes for workers. These programs are distinctive in their focus on high-growth sectors of the economy such as information technology, health care, and advanced manufacturing. The federal government should substantially increase investment in these programs, focusing on replication and scaling of programs with track records of success. For our most disadvantaged students and workers, who sometimes have difficulty qualifying for participation in these programs, additional supports and “on-ramp” programs should be considered.

We should also strengthen the “connective tissue” of supportive services. Education, training, and employment systems are decentralized, and the bewildering array of options can overwhelm workers who are juggling busy lives on top of their training needs. Barriers related to transportation, child care, and mental health often cause program participants to exit programs early. This is a lost opportunity. Moreover, the evidence shows that counseling and supportive services, both of which are integral to the sectoral strategies mentioned above, substantially increase program completion and labor market success. Investments in support services for post-secondary training participants such as community college students and displaced workers can yield high returns.

A third critical need is for innovation in the nation’s workforce data infrastructure. Workers are pressured by technological change and automation, which makes it critical to modernize our education and training systems to keep up with change. We need better information about which jobs are growing and which programs are effective at developing needed skills. For example, one member of our group is developing a framework for decentralizing regional labor market information systems that will help states and regions develop deep and agile data systems for measuring program performance as well as changing skill and employment needs.

Finally, the evidence of “what works” in training and workforce development programs is remarkably sparse. Even if political will existed for a full-scale, far-reaching reform of WIOA, community colleges, and other elements of our workforce system, it would be imprudent, based on what we know, to recommend a one-size-fits-all model for all regions and priority industries.

In light of this uncertainty, we need strategies that unleash innovation at the state and regional levels and among industries. Part of the answer to this challenge is providing state and local officials substantial flexibility in testing new program structures and models that bridge public, private, and nonprofit institutions; are responsive to fast-changing demand patterns; and meet the differing needs of populations ranging from English language learners to working adults to the formerly incarcerated. Such experiments deserve more financial and implementation support from the federal government, opportunities for administrative flexibility, and comprehensive evaluation to help inform future rounds of system reform.