State of the Paper Ceiling 2026: Moving Toward a Better Labor Market
Workers move through the labor market like drivers move through a highway system — taking on and off ramps, reading signs and signals, navigating toward opportunity. But the infrastructure that guides that journey has been quietly breaking down. This is a report about what it will take to rebuild it right.
State of the Paper Ceiling 2026: Moving Toward a Better Labor Market
Dead End Streets and Broken Signals
Infrastructure shapes access.
Like roads, labor market infrastructure can move people toward opportunity or route them away
from it. And like all infrastructure, it reflects the priorities of the people who designed it.
For the more than 70 million workers who are skilled through alternative routes — STARs — that
design was not built with them in mind.
The numbers tell a stark story. From 2000 to 2020, STARs lost access to 7.5 million critical jobs — positions such as administrative assistant and sales representative that had once served as reliable gateways to higher roles and better wages.
The Labor Market Has a Design Problem
These pathways didn't erode on their own. They were the casualty of design decisions that used degrees as a proxy for skills — embedding pedigree in hiring processes where skill should have been the measure. These cumulative decisions remade the labor market in a way that limited STAR mobility and unnecessarily restricted employer access to real talent.
Design Problems Can Be Fixed
The good news is that because people design the labor market, they can redesign it. When employers use signals that are better reflections of skills — dropping degree requirements, rewriting job descriptions around skills, adopting assessments that make worker capabilities visible — jobs reopen to STARs and pathways widen. We know this works because we have seen it happen:
In 2020, the data shifted. The two-decade decline in opportunity for STARs began to reverse, and STARs started regaining share in roles they had lost — yielding 783,000 jobs.
Better Design Leads to Better Outcomes
This improvement was not random. It was concentrated where design was being deliberately changed: among employers adopting skills-based hiring, and most sharply within Opportunity@Work's employer networks and public sector cohorts.
The Evidence Is Building
More jobs are open to STARs than at any point in the last decade — proof that redesign works at scale.
Year-over-year increase in jobs open to STARs from employers in our network, a total of almost 600,000 jobs.
states are now implementing skills-based practices — a signal that the movement has reached policy.
STARs have experienced better hiring outcomes in O@W employer networks than the labor market. 52% of STARs that started a new job in our network experienced upward mobility (vs 39% overall).
of STARs starting new jobs in the broader labor market achieved mobility in the most recent year.
of STARs starting new jobs in our network achieved mobility in the most recent year.
AI Will Amplify Our Intentions
In 2026, a technological transition is well underway and we must shape it. Deployed without intention, AI will do what unexamined systems have always done: encode historical exclusion, accelerate existing inequities, and bury them inside an algorithm where they become harder to see and harder to challenge. The design problem doesn't disappear with new technology. It scales.
With Intention, AI Can Open Opportunity
Whether AI reshapes job pathways toward access or exclusion and whether AI-driven hiring tools surface worker talent or suppress it are not questions the technology will answer on its own. They are design questions. And design questions have designers: people and institutions making choices right now about what this infrastructure will do, and who it will serve.
The Design Is Up to Us
Labor market infrastructure doesn't build itself. It is designed, piece by piece, by people with choices to make. Five key stakeholders have the power to shape what comes next — and each has a critical role to play. The question is not whether the infrastructure will be designed. It will be. The question is whether we design it with intention — and with everyone in mind.