Centre for Skill Development · WAC Annual Report · 2025 edition

Skills Gap Report 2025

The skills gap is less a shortage of talent than a failure to measure, recognize and build it

128 pages12 chapters272 references36 chartsPublished 12 October 2026Free PDF
Get the report
Cover of Skills Gap Report 2025

The skills gap is real but narrower, more specific and more self-inflicted than headlines suggest. Drawing on about 270 studies, official statistics and surveys, this first WAC Skills Gap Report finds genuine shortages in care, trades, energy and AI, weak and stagnant foundational skills, and widespread under-use of talent. Much of the gap lies in how employers measure, recognize and build skills. The report introduces the WAC Skills Readiness Framework and ten signals to watch in 2026-2027.

The report in numbers

72%Employers reporting difficulty finding talent, 2026
26%OECD adults at Level 1 or below in literacy
19%Employment gap, US 22-25s in AI-exposed jobs
39%AI users who received employer AI training
1 in 700Max share of US hires changed by dropping degree rules
14-39%Earnings gains from sectoral training in RCTs
52%US graduates underemployed a year after graduating
2.6xEU learning rate, graduates vs least educated (WAC calc.)

Key findings

  1. Seventy-two percent of employers in 41 countries report difficulty finding talent in 2026, down from a 77 percent peak in 2023 but roughly double the 2014 level.
  2. Official data show shortages easing: UK skill-shortage vacancies fell from 36 percent of all vacancies in 2022 to 27 percent in 2024, and US job openings fell to 7.1 million by August 2026.
  3. Under-use of skills rivals under-supply: 26 percent of workers in the OECD Survey of Adult Skills say they are over-skilled for their job, against 10 percent under-skilled.
  4. Foundations are weak and not improving: 26 percent of adults in OECD countries score at Level 1 or below in literacy and 29 percent in adaptive problem solving, and only Finland and Denmark significantly improved adult literacy over the decade.
  5. Employment of 22- to 25-year-olds in AI-exposed US occupations is 19 percent below where it would have been had it kept pace with less-exposed peers, while economy-wide effects remain small.
  6. Only 39 percent of workers using AI at work had received AI training from their employer, while 66 percent of leaders said they would not hire someone without AI skills.

The report has 12 key findings in all, 12 chapters of evidence, an original framework and recommendations for five audiences.

Inside the book

  1. Chapter 1

    What the skills gap is, and how it is measured

    One phrase covers shortages, proficiency gaps and mismatch; each is measured differently, and the method often decides how large the gap appears.

  2. Chapter 2

    The global picture: what employers, workers and tests show

    Employer complaints doubled in a decade and are easing only slightly, while tests show adult basic skills stagnating in rich countries and lagging from childhood in poorer ones.

  3. Chapter 3

    AI and the reshaping of skill demand

    Generative AI lifts novices, squeezes entry-level hiring in exposed jobs and pays a premium for AI skills, while employer reskilling lags its rhetoric.

  4. Chapter 4

    The skills in shortest supply

    Real shortages cluster in AI, cybersecurity, trades and health care, while the human and foundational skills beneath them are both prized and thinly spread.

  5. Chapter 5

    The skills gap around the world: a regional analysis

    From graduate surpluses in Asia and North Africa to demographic shortages in Europe and East Asia, the gap takes different shapes and needs different responses.

  6. Chapter 6

    Sector by sector: where the gaps bite

    Shortages in care, trades, energy and teaching are driven by demography, pay and slow pipelines, while technology and finance face changing entry-level work.

  7. Chapter 7

    Skills-based hiring: promise and practice

    Employers and governments have stripped degree requirements from postings, but hiring has barely moved; valid assessment, not edited job ads, is what changes outcomes.

  8. Chapter 8

    What works: evidence on closing skills gaps

    Randomized trials favor sector-focused training tied to employers; broad vouchers, generic training and self-paced online courses have weaker records.

  9. Chapter 9

    Who gets to upskill: workers, learners and inequality

    Adult learning flows to those who already have the most skills, in the largest firms, and in their prime working years; money alone has not reversed that.

  10. Chapter 10

    What the gap costs

    Trillion-dollar estimates of the skills gap are model scenarios, not measured losses; the firm and worker evidence shows real but smaller and more specific costs.

  11. Chapter 11

    The WAC Skills Readiness Framework

    A practical tool, synthesized from this report's evidence, for finding where a skills gap really lies and what universities, employers and governments should do next.

  12. Chapter 12

    Outlook 2026-2027: what to watch

    Ten signals will show whether the skills gap is narrowing, shifting or widening, and what evidence would change the picture.

A chart from the report

Firms that dropped degree requirements: did hiring change?

Share of firms studied

0%10%20%30%40%50%Leaders (hiring changed)37%In name only45%Backsliders18%
Firms that removed BA requirements from postings, 2014-2023, classified by subsequent hiring of non-degree workers; about 11,300 roles at large US firms. Source: Burning Glass Institute and Harvard Business School Project on Managing the Future of Work (2024).

One recommendation for each audience

For universities

Publish graduate outcomes by program, measured as employment in jobs that use the degree, and tie program growth to that evidence rather than to enrollment demand; 52 percent of US graduates start underemployed and most who start that way stay there.

For employers

Before declaring a shortage, check pay against the market, audit applicant-tracking filters and compare training spend with hiring cost; in cybersecurity, budget rather than candidate supply was the most cited cause of shortages.

For edtech and training providers

Build for completion and verified outcomes, not enrollment: completion in HarvardX and MITx courses fell to 3.13 percent, against 46 percent among learners who paid for verified certificates.

For policymakers

Report skills shortages alongside vacancy duration, wages and training data, and fund repeated direct tests of adult skills, since employer surveys alone can overstate and misdiagnose gaps.

For learners and workers

Choose programs by field and graduate outcomes, not prestige alone: five years after graduation, underemployment is 23 percent for health graduates but 57 percent for general business fields.

This report is desk research: a synthesis of about 270 published studies, official statistics, employer and workforce surveys and management-journal analyses. WAC ran no survey or trial for it. Every chapter was checked against its sources by an independent fact checker before publication. The photographs are AI-generated illustrations. Send a correction

From the same centre

See what the Centre for Skill Development works on: its white papers, Spotlight analysis and the news in its field.

Visit the centre