Business professionals reviewing an application form during a meeting.

Why Skills Are Becoming the New Measure of Workforce Readiness

ByAmanda Coolidge

In 2022, Delaware’s Department of Human Resources went through its Family Service Specialist job postings and replaced the bachelor’s degree requirement with a line about relevant work experience. Applications went up 575 per cent.

That number says less about the degree requirement than about who was already there: people doing adjacent work, capable of the job, invisible to the hiring process because the posting was filtering on the wrong thing. A degree requirement doesn’t just screen out people who lack a degree. It screens out people who have the skill but acquired it some other way (on the job, through a parent’s business, through years in a related role, etc.), and have no credential to show for it. Delaware didn’t lower its bar, rather it stopped measuring candidates with an instrument that had nothing to do with whether they could do the work.

Maryland made the same call that year, and 24 more states have since followed, stripping degree requirements out of public postings one agency at a time. IBM had already been running the same experiment for the better part of a decade. More than half of its U.S. postings no longer ask for a degree, and about a fifth of its American hires now don’t have one. Applications from underrepresented groups climbed roughly 63 per cent once the requirement came off, the same mechanism as Delaware’s number, just measured from the demand side instead of the supply side. 

Credentials are proxies, not measurements

A new report from the Organisation for Economic Co-operation and Development (OECD), A Skills-First Labour Market (2026), gets at why this filter is so leaky in the first place. Using its own Survey of Adult Skills, the OECD found that years of schooling explain only about a fifth of the variation in a person’s actual skill proficiency. In plainer terms, two people who completed the exact same credential can walk away with very different real capability, depending on what happened during that credential, how rigorous it was, how much of it they engaged with, and what they did before and after it. A degree was always a proxy standing in for capability, never a direct measurement of it. 

A mismatch in both directions

More than a third of workers across OECD countries, and over 40 per cent in the U.S., Canada, and Germany specifically, say their skills don’t match what their job requires. At the same time, more than eight in ten employers in Japan, Germany, the Slovak Republic, Greece, and Ireland say they can’t find candidates with the skills they need. 

Both can be true in the same labour market because the sorting mechanism is broken in both directions at once. It screens out people who could do the job, which shows up as the worker-side mismatch, while failing to reliably identify who among the people it lets through can do it, which shows up as the employer-side shortage. The right instrument would fix both problems at once. A credential-based one can’t, because it was never measuring the thing either side needs to know.

The problem gets harder when skills keep changing

The World Economic Forum’s Future of Jobs Report 2025, drawing on more than 1,000 employers representing 14 million workers, puts a clock on how much worse this gets. Employers expect 39 per cent of the core skills their workforce needs today to be different by 2030. 

A two-year program built around today’s job requirements, for instance, graduates its first cohort into a labour market where four in ten of the skills that program assumes are already different. A job requisition written last year, reused because rewriting requisitions is tedious, is recruiting for a version of the role that no longer exists without anyone noticing. An annual review cycle built around whether someone completed a certification has no way to track a skill that changed mid-cycle.

Skills-first approaches are already showing results

LinkedIn found that recruiters using skills data to fill roles make a successful hire 60 per cent more often than those relying on credentials alone. Deloitte found organizations built around skills-based practices are 63 per cent more likely to hit their performance targets, and retention moves the same direction. 

None of this means credentials are worthless, or that every institution should tear up its programs tomorrow. A degree still bundles things a skills list doesn’t always capture well, sustained effort over years and exposure to material a learner wouldn’t have chosen on their own. The honest version of this argument is narrower than “credentials are dead.” Credentials were never built to tell an employer, precisely and currently, what someone can do, and that gap between what a degree signals and what a job requires only widens as the pace of change picks up.

The harder problem isn’t persuading anyone that skills matter. Employers, governments, and students largely believe that already. The harder problem is building the infrastructure to describe a skill the same way twice and prove someone has it in a way a stranger can trust. That description also has to stay current once the skill itself changes, which it inevitably will. 

References


Lead photo by Kampus Production