Overqualified: graduates in jobs that do not need the degree
42 recent papers study skills mismatch, overeducation and graduate employability, up from 24 two years ago, mostly in economics. Graduates who find work often find it below their level, and generative AI is changing the entry-level tasks they would learn on. Does starting overqualified leave a lasting mark?
What prompted this#
Since October 2025, 42 papers carry a skills-mismatch, overeducation, job-mismatch or graduate-employment keyword, up from 30 in the year before and 24 two years earlier: "skills mismatch" on 19, "overeducation" on 8, "graduate employability" on 8, "graduate employment" on 8. Twenty-six appeared in economics journals and 10 in management journals, mostly on management education. Over three years they sit next to gender differences (9), university graduates (6), underemployment (4), higher education expansion (4), returns to education (4) and school-to-work transition (4); only 4 of 96 carry an AI keyword. A related note is about graduates without jobs; this one is about graduates whose jobs do not need their degree.
Questions#
- Do graduates who start overqualified catch up with their peers, or does the mismatch persist for years?
- Is mismatch rising by field of study as higher education expands, and is it worse for women?
- Does generative AI raise or lower the skill content of graduate entry jobs?
- Self-reported mismatch and job-requirement measures (occupation codes) often disagree: by how much, and which tracks wages?
Data that could help#
The OECD Survey of Adult Skills (PIAAC, second cycle with 2023 data), graduate tracer surveys such as the UK's Graduate Outcomes, and job advertisements with stated education requirements.
Next step#
Compare three overeducation measures (self-report, occupation requirements, realized matches) in the two PIAAC cycles for the countries in both, and check whether their trends agree.