Whenever I complain to my executive coach about a problem inside my organization, his go-to response is both irritating and insightful: “That’s your system.”
I might be describing a missed target, a persistent challenge, or something that feels impossible to fix, but his point is always the same. The system is producing exactly what it was designed to produce. I need to stop looking for someone to blame and start looking at the design.
My job is to get schools, employers, and governments working together to move more young people into good jobs. And we run into the same design flaw over and over. It’s that America’s education and workforce system was built to select people out, not move everyone up.
This problem is worsening. Research from the International Labour Organization found that in 2025, youth unemployment rose to 12.4%, up from a two-decade low near 12% in 2023. This was the first reversal in years, equal to 67 million young people out of work worldwide. Some of that is AI reshaping entry-level work: 6.1% of jobs held by young people sit in occupations highly exposed to AI-driven change, concentrated in the clerical and administrative roles that used to be someone’s first rung on the ladder. At the same time, employers say they can’t find enough skilled workers to keep up with reshoring, the energy transition, and the physical buildout AI itself requires. Both ends of the pipeline are breaking down together.
And the math is about to get harder. The U.S. Bureau of Labor Statistics projects that the number of 16- to 24-year-olds in the U.S. labor force will fall by 2.04 million—nearly 10%—between 2025 and 2035, while the number of high school graduates is projected to decline 13% between its 2025 peak and 2041. Employers and colleges are about to compete for a smaller generation of young people, even as both continue to operate systems designed to reject large shares of them.
A SYSTEM BUILT TO SELECT, NOT SUPPLY
For decades, employers had the luxury of choosing among an abundant worker supply, so the whole system organized itself around sorting. Admissions got tougher and credential requirements crept upward. Degrees became a stand-in for capability. None of that was designed to create a shortage, but to create confidence.
Confidence isn’t supply, though. Worldwide, 20.4% of young people are not working, in school, or training. Rates for young women are often double those for young men. While we have plenty of people, we don’t have enough pathways.
A system built for selection also loses people along the way, and those losses add up fast. Enrollment gets treated like supply. A program announces 10,000 participants and everyone assumes 10,000 future workers are on the way. In reality, people drop out at every stage for multiple reasons: scheduling conflicts, cost, a certification they never finished, a hiring process that screens out someone who’s already proven they can do the job. By the time someone becomes a productive employee, the original cohort has usually shrunk dramatically. That shrinkage is a design choice, not a talent shortage.
A shrinking generation leaves far less room for waste. With fewer young people entering the labor market, every unnecessary credential, unpaid internship, inaccessible schedule, and automated screen removes capacity the economy cannot easily replace.
Even so, most companies still treat workforce development as someone else’s job. Schools are supposed to prepare workers, while governments are supposed to fund the gaps. Then companies hire from whatever pool shows up. And when that pool falls short, the response is usually to complain about it rather than invest in it.
BUILD BACKWARD FROM DEMAND
So what happens when a company builds its pipeline backward from real hiring needs instead of forward from a generic curriculum? With FedEx, we built Jóvenes con Entrega, or “youth who deliver.” We started with FedEx’s logistics roles in Mexico, mapped the skills those jobs required, brought in FedEx’s managers to help design the training, and built mentorship and internship pathways directly into technical high schools.
More than 34,000 young people have graduated so far. Forty-eight percent are women. Just over half secured jobs within four to six months, 15% above target. A study of 4,000 graduates found the program turned FedEx’s $1.5 million investment into nearly $4.5 million in added income for participants, a 204% return.
FedEx’s return isn’t an outlier. Every $1 invested in youth skills generates $10–15 in economic growth in developing countries, per UNESCO’s Education for All Global Monitoring Report—the kind of payoff most CFOs would move on instantly if it showed up anywhere else on the balance sheet.
That’s what happens when training starts from real hiring demand instead of hoping jobs will follow a curriculum. Most workforce programs still get built the other way around, and it’s why they lose people at the transition into work, not just at enrollment.
This requires a reframe rather than a bigger budget. Instead of asking how many people you can train, ask how many you can move into work. Audit your pipeline stage by stage and find out where people are falling out. Co-design training backward from your real hiring needs, the way FedEx did. And measure placement and independent performance, not completions, because a certificate means nothing if the person never makes it to the job.
Workers power economies. The companies that continue filtering as though talent were abundant will spend the next decade competing over a shrinking pool. The next era of growth will belong to those willing to enlarge it by building up the people the old pipeline was designed to leave behind.
Christina Sass is president and CEO of International Youth Foundation.
