The 25-Year Detour

Pop quiz: Who said this, and when? 

“We have the best system of higher education in the world, but we are the only advanced country that basically has no system for helping all of our young people…have a smooth transition from school to work…We propose a better system…a new kind of education and training connecting the classroom and the workplace, removing the artificial distinction between the academic and the practical… We have an opportunity to do something that Americans have resisted for too long, which is to merge instead of keep divided our notion of vocational education and academic education…A B.A. degree should not and must not be the only ticket to a good job and a good livelihood…”

I can’t fault you for guessing it’s a quote from Thriving After High School: Community, School, Service, and Work, a report released last week from the American Academy of Arts & Sciences’ Commission on Education After High School. And I won’t blame you if you thought about Beyond silos: A state policy playbook for building integrated education-to-career systems, a new series of briefs from the Christensen Institute, also released last week. A lot of smart people are working hard to reimagine a system that better connects learning and work these days. 

But the quote above isn’t actually from 2026. That’s President Clinton in 1994, just months before he signed the School to Work Opportunities Act (STWOA) into law. Passed at a moment when the economy was growing, but Americans were increasingly anxious about jobs, wages, and the changing nature of work (sound familiar?), the Act directed $1.85 billion to state and local partnerships to lay the groundwork for a new national approach to helping young people make the transition successfully from education into careers.

STWOA required participating programs to integrate academic learning, work-based learning, and ‘‘connecting activities’’ to purposefully bridge the two—capacity to provide things like placement services, technical assistance to schools and employers, and to collect information on student outcomes (akin to intermediary functions in today’s parlance). By 1998, the effort had reached all 50 states, D.C., and Puerto Rico. Survey data published in 2000 found that STWOA funding supported more than 770 local partnerships encompassing nearly 18 million students, more than 36,000 schools, 178,000 employers, and 2,600 postsecondary institutions. As the report noted, ‘‘States and local partnerships have shown demonstrable progress toward building systems and designing and implementing service strategies that are affecting the ways in which students are educated across the country.’’ 

But on October 1, 2001, twenty-five years ago today, that landmark investment quietly sunset. Three months later, President Bush signed No Child Left Behind, ushering in an era of test-based accountability and creating the conditions for the “College for All” movement to take hold. With no plan to sustain the STWOA’s momentum and no place for it in the new accountability regime, federal policy reverted to supporting pieces of the system—career and technical education, workforce development, higher education, apprenticeships—without building the critical linkages across them. Today, they remain largely siloed, with fewer and fewer staff to maintain bridges between them. If we hadn’t taken a detour 25 years ago, would this still be the case?

It’s hard to know, of course. STWOA was not perfect, nor was it a  failure. But it didn’t succeed in producing the ambitious national system that its architects imagined. So, what happened—and what lessons can its fate offer today’s career pathways movement? Here are four:

1. Measure what you mean.

Perhaps the most ambitious part of STWOA was its aspiration to create a “universal, high-quality school-to-work transition system.” But what, exactly, would that system look like? In 1994, no one was quite sure. To a large extent, we still aren’t. 

As STWOA’s lofty vision was translated into a federal grant program, states were asked to turn an  abstract goal into things they could propose, fund, implement, and measure on a roughly five-year timeline. Grant requirements emphasized easy-to-tally metrics: partnerships formed, programs launched, students enrolled, employers engaged, credentials earned. These measures matter, of course, but they don’t do much to encourage systems change efforts, let alone assess them.  

To be fair, evaluations of STWOA did seek to assess system change, but they reveal the limits of what could be measured and accomplished in such a short time period. STWOA’s evaluation apparatus could document a great deal of activity and implementation, but without a clear framework to define or measure system change, it was not possible to assess the extent to which states were adopting policies and implementing structures to operationalize that goal. In the end, most weren’t. What gets measured, as they say, gets done. 

The lesson? We can’t change systems simply by funding and tracking programs. Today’s career pathways movement needs to be more clear about the system characteristics and outcomes we are trying to achieve, and then design funding and accountability structures that reinforce those goals. Program outputs matter, and program outcomes more so. We still need better data systems to collect them. But if our goal is lasting transformation, we can’t sleep on the system change metrics. 

2. Experiment, but learn together.

Another hallmark of the STWOA was the flexibility it provided to states and local partnerships to define and design their own school-to-work systems. On one hand, this made sense: different communities, institutions, and regional economies required unique solutions; and then, as now, no one wanted Washington to prescribe a one-size-fits-all solution. 

On the other hand, however, the proliferation of approaches made it difficult to understand what was working. State and local grantees generated a substantial body of experience and evidence, but it was diffuse. A 1998 Department of Labor review found that states had wide discretion over what data to collect, how to define key terms, how long to follow participants, and which measures to track. Most were building evaluation capacity and data systems largely from scratch, leading the report to predict (correctly, it turned out) “wide variation” in states’ evaluation approaches and findings. 

STWOA yielded a lot of activity, but the law’s flexible architecture made cumulative learning difficult. Across a robust collection of implementation studies and evaluations, results on STWOA’s effectiveness were mixed. Ultimately, the lack of clear conclusions about what worked and what didn’t made it challenging to advocate for further investment.  

The lesson? States need room to innovate, but they also need enough common language and infrastructure to learn from one another. Common definitions, shared measures, stronger data systems, and sustained cross-state learning efforts can make experimentation cumulative rather than simply additive. States don’t need to build identical systems, but we do need to get better at learning together about what makes those systems work.

3. Employers need systems, too.

STWOA understood that employers had to be part of the system, and by 1998, partnerships were reporting nearly 180,000 employers involved in their work. But breadth wasn’t depth. Though the law placed its greatest emphasis on intensive work-based learning opportunities, student participation was rare. By programs’ own conservative estimates, just 8% of students job shadowed and 3% completed internships in the 1997 school year. 

The economics explain why. Hosting a student for a morning is cheap. Hosting one for months means recruitment, matching, supervision, and paperwork, with uncertain return. Most participating employers said they were motivated by community contribution or public relations rather than true business needs, and most non-participants didn’t see a financial reason to join. 

Employers in the 90’s asked the same questions employers ask of today’s work-based learning models: Who bears the cost? Who coordinates schedules? Who is liable? What’s in it for the business? STWOA anticipated these challenges by requiring “connecting activities” like matching students with employers, providing mentors and technical assistance, and designing work-based learning opportunities. But these functions lived largely within local partnerships and were dependent on staff capacity and grant-funding. Relationships were forged, but systems largely were not. 

The lesson? Employer engagement can’t be a series of increasingly complex invitations for individual businesses to participate. We need a more systemic approach to bringing employers into career pathways—one that gives them clear incentives to participate, provides the infrastructure to make participation feasible, and creates mechanisms for turning cursory engagement into deeper, sustained relationships. Otherwise, we risk spending enormous amounts of time and energy solving the same problems one employer at a time, and struggling to make a broader case for their role as true partners in a career pathways system that delivers mutual benefit. 

4. Plan for longevity, not launch.

STWOA was designed from the jump to “provide seed capital to States and localities for developing and implementing comprehensive STW Opportunities systems.” The logic was clear, but states had very little time to build those systems through STWOA, and even less time to make the case to state legislatures for sustaining them. Most STWOA grantees had just five years to dramatically change systems, incentives, behavior, and make a case for continued support; it was not enough time. Compounding the challenge, STWOA’s federal funding began winding down just as states entered a period of fiscal stress. The 2001 recession put pressure on state budgets, making it harder for states to replace temporary federal grants with ongoing funding. But the deeper challenge was structural: the most critical infrastructure STWOA was trying to create—cross-agency coordination, work-based learning at scale, intermediary capacity—didn’t fit neatly into state budgets. To some extent, it still doesn’t.

The lesson? System building requires time, patient capital, and a deliberate plan for institutionalizing the work. It also requires giving the connective infrastructure a prevalent and permanent home in policy and budgets, rather than supporting it with short-term grants, public or private. In today’s career pathways movement, much of the support for intermediaries has come from philanthropy, which has a notoriously short attention span. Funders in some cases have succeeded in passing the baton to states, but with elections looming and new leaders arriving, that progress could soon be at risk. Systems change is a long game. The moment to stick with it is when the novelty wears off, and in many places, we’re nearly there. 

Twenty-five years after STWOA sunset, we’re back on the same track: trying again to build a system that better connects education and work to expand opportunity in an uncertain economy. But this time, we have the benefit of knowing where the last journey took us. These are just four of the many lessons we can learn from STWOA’s legacy. Let’s take them to heart so this time we have a better chance of staying on course.