Job-Ready Today, Job-Obsolete Tomorrow? Vocational Training as a Substitute for University in Europe

In September 2025, Spain's Ministry of Education announced a record. The country had enrolled 1,218,347 students in Formación Profesional (FP) for the 2025-2026 academic year, the ninth consecutive year of growth and roughly 36% more students than five years earlier. In the same week, Eurostat reminded everyone that Spain also holds the European Union's worst overqualification rate: 35.0% of Spanish workers with a tertiary degree are employed in jobs that, by ILO occupational classification, do not require one. The EU average is 21.3%.

The narrative writes itself: too many graduates, not enough graduate-level jobs. So Spain, and to varying degrees France, Italy, Portugal and Greece, has begun pushing students earlier and more decisively toward vocational tracks, often as an explicit substitute for university. Brussels supports the move, the OECD applauds it, and CEDEFOP measures it. The German dual system has gone from a national peculiarity to a continent-wide aspiration.

But the empirical economics of substituting vocational training for university is more uncomfortable than the policy slogan suggests. Vocational education does deliver the gains it promises. It just hands a non-trivial part of them back, twenty years later.

The Substitution Hypothesis

The policy case for VET-as-substitute rests on three empirical observations, each well documented.

First, the gap in school-to-work transitions is large. In Spain, 51.1% of Grado Superior graduates from the 2020-2021 cohort were affiliated to Social Security one year after graduation, and that number climbs across the second and third years (Ministerio de Educación, FP y Deportes, 2025). University graduates, in the same horizon, fare worse. Eurostat shows that across the EU, employment rates at age 20-34 are systematically higher for recent VET graduates than for recent graduates of general upper-secondary education.

Second, the wage signal looks competitive at entry. Across Europe, hourly wages of VET graduates at labour-market entry are not dramatically below those of recent university graduates, and in technical specialities the rank ordering can invert. The headline numbers from sectoral surveys in Spain, with entry-level salaries of around €22,000 for Grado Superior graduates versus €26,000 for university graduates (Adecco-Infoempleo data, reproduced in the Spanish press), are not the whole story, but they are not a rout either.

Third, the causal evidence on workplace-integrated training is convincing. The cleanest recent estimate is Bentolila, Cabrales and Jansen (2023), which uses matched education and Social Security records for the first three cohorts of dual FP graduates in the Comunidad de Madrid, with a distance-based instrumental variable to handle student self-selection. They find that dual VET graduates accumulated 27% more days of employment and earned 32% more during the first twelve months after graduation than school-based FP graduates in the same field and year. They also find no significant effect on hourly wages or on transitions to indefinite contracts, confirming that the gain is mostly on quantity (more employment) rather than quality (better jobs).

For policymakers facing a youth unemployment rate that briefly touched 56% in Spain in 2013 and is still uncomfortably high across southern Europe, these three observations look like a complete case. Less time at university, more time on a shop floor, better outcomes.

The argument is not wrong. It is incomplete.

The Lifecycle Trade-off

The seminal piece of evidence on what happens later is Hanushek, Schwerdt, Woessmann and Zhang (2017), published in the Journal of Human Resources. Using the International Adult Literacy Survey covering 11 OECD economies, the authors estimate a difference-in-differences specification that compares employment rates of vocational and general-education workers at different ages. The picture they document is sharp: at 25, the vocational worker is more likely to be employed. By 50, in the same countries, she is less likely to be employed. The cross-over is largest in countries with strong apprenticeship systems (Germany, Austria, Denmark, Switzerland), exactly the systems being held up as models.

Subsequent replications have confirmed the basic pattern. Korber and Oesch (2019), exploiting Swiss panel data, document that for men, vocational education produces a small lifetime earnings advantage; for women, general education dominates the earnings profile from about age 30 onward. Forster, Bol and Van de Werfhorst (2016), using PIAAC data, confirm the cross-over pattern in cross-national analysis: vocational workers begin with higher employment probabilities than their general-education peers, but the differential reverses later in the career.

The reason is structural. Apprenticeship-style VET transmits specific human capital: how a particular machine works, how a particular workflow is built, how a particular firm's product is assembled. Those skills are immediately productive, which is precisely why employers hire VET graduates faster. But they depreciate as technologies, workflows, and product mixes evolve. General education transmits transferable cognitive skills, slower to monetise on entry but more resilient to occupational displacement. In the framework of Acemoglu and Restrepo (2019), where technical change continuously redraws the boundary of tasks done by humans, the worker with broader cognitive skills can recompose her task bundle when her old one disappears. The vocational worker often cannot.

This is not a small effect, and it is not symmetric across countries. The trade-off is largest where the VET system is strongest, because that is where the specialisation is sharpest.

Spain Is the European Test Case

Spain matters here because it has, over the past decade, built the most lopsided tertiary system in the EU. OECD's Education at a Glance 2025 reports that 39% of Spanish young adults enter higher education through short-cycle programmes (overwhelmingly Grado Superior of FP), versus 16% across the OECD and 10% in the EU25. Eurostat confirms the consequence: 24.7% of all Spanish tertiary students are in short-cycle programmes, comfortably the highest share in the EU.

At the same time, 52.6% of Spaniards aged 25-34 hold a tertiary degree (versus the EU average of 44.1%). Spain is simultaneously running a mass-university experiment and the EU's largest short-cycle vocational sector. The result is the overqualification number with which we began.

Table 1 makes the picture concrete.

Table 1. Tertiary attainment and overqualification of tertiary graduates, selected European countries, 2024.

Country	Tertiary attainment, 25–34 (%)	Overqualification rate (%)
Spain	52.6	35.0
Cyprus	62	28.2
Luxembourg	60	4.7
Lithuania	57	n/a
Netherlands	55	n/a
Sweden	54	n/a
France	52	n/a
Belgium	50	n/a
EU27 average	44.1	21.3
Italy	31	n/a
Greece	n/a	33.0
Czechia	n/a	12.8
Croatia	n/a	12.6

Sources: Eurostat, EU's employment rate reached almost 76% in 2024 (15 April 2025) for overqualification rates; Eurostat, 43% of EU's 25-34-year-olds have tertiary education (27 May 2024) for tertiary attainment shares; OECD, Education at a Glance 2025, for the Spanish 52.6% figure. Cells marked n/a denote that the same Eurostat release does not provide the matching figure; the omission is not a measurement.

The pattern that emerges is not 'tertiary attainment causes overqualification'. Luxembourg, Cyprus and the Netherlands all sit above the EU tertiary average, with overqualification at or below it. What distinguishes Spain, and to a lesser extent Greece, is the combination of high tertiary attainment, a large short-cycle vocational sector, and a productive structure that has not generated enough high-skilled occupational demand to absorb either output. The policy debate then frames vocational training as the solution to the overqualification problem, when in fact both numbers grew together over the past decade.

What the Trade-off Implies for Policy

There are three implications worth being explicit about.

First, the appropriate comparison between vocational training and university is not at age 25. It is over the working life. Studies that stop at the school-to-work transition will systematically overstate the relative return to VET, and studies that look only at age 50 will overstate the relative return to university. Both are real; neither is the whole story.

Second, dual systems, those that combine firm-based and school-based training, appear to dominate purely school-based VET on the dimensions where VET is supposed to deliver. The Bentolila, Cabrales and Jansen (2023) results are clear on that. But the dual systems are also the ones where the lifecycle reversal is sharpest in Hanushek et al. (2017). The technology that makes a workplace apprenticeship valuable at 22 is the same technology that risks making the apprenticeship obsolete at 50.

Third, the recent move in Spain to expand FP at the expense of university, increasingly framed in policy circles as a corrective to overqualification, is in equilibrium betting on a particular life-cycle distribution of human capital. It is buying employment at 25 by selling some employment at 55. That trade may be the right one, given the youth unemployment Spain still carries; whether it is the right one in the long run depends on how successful the system is at building bridges back into transferable skills.

A complementary policy, less politically convenient but more empirically defensible, is to invest in the general component of VET (literacy, numeracy, transferable cognitive skills, and bridging routes back into tertiary education), precisely to flatten the lifecycle reversal. The Spanish 2022 reform of FP (Ley Orgánica 3/2022, de Ordenación e Integración de la Formación Profesional) explicitly embeds these bridging routes; whether it is enough to close the gap that Hanushek and coauthors document remains an open empirical question for the next decade.

For a young person in Madrid, Lyon or Athens making a decision in May 2026, the relevant question is not 'does VET get me a job faster than university?'. It plainly does, in most fields. The relevant question is: 'does it get me a job for the next forty years?'.

The European evidence does not say no. It says the answer depends on how technology evolves, on how general the vocational track is in practice, and on whether the educational system makes it possible to come back later for the cognitive skills that depreciate more slowly. The interesting empirical agenda is not whether VET works, but for whom, in which sector, and for how long.

Spain, with the largest short-cycle vocational sector in Europe and the worst overqualification rate, is the laboratory. The rest of Europe will read the results.

References

• Acemoglu, D., & Restrepo, P. (2019). Automation and new tasks: How technology displaces and reinstates labor. Journal of Economic Perspectives, 33(2), 3–30.

• Bentolila, S., Cabrales, A., & Jansen, M. (2023). Does dual vocational education and training pay off? CESifo Working Paper No. 10762. Munich: CESifo.

• European Commission. (2024). Education and training monitor 2024: Spain. Luxembourg: Publications Office of the European Union.

• Eurostat. (2024, May 27). 43% of EU's 25-34-year-olds have tertiary education [News article]. https://ec.europa.eu/eurostat/web/products-eurostat-news/w/ddn-20240527-1

• Eurostat. (2025, April 15). EU's employment rate reached almost 76% in 2024 [News article]. https://ec.europa.eu/eurostat/web/products-eurostat-news/w/ddn-20250415-1

• Forster, A. G., Bol, T., & Van de Werfhorst, H. G. (2016). Vocational education and employment over the life cycle. Sociological Science, 3, 473–494.

• Hanushek, E. A., Schwerdt, G., Woessmann, L., & Zhang, L. (2017). General education, vocational education, and labor-market outcomes over the lifecycle. Journal of Human Resources, 52(1), 48–87.

• Korber, M., & Oesch, D. (2019). Vocational versus general education: Employment and earnings over the life course in Switzerland. Advances in Life Course Research, 40, 1–13.

• Ministerio de Educación, Formación Profesional y Deportes. (2025, November 26). Aumenta el porcentaje de alumnado de Grado Medio y Grado Superior que trabaja al año de graduarse [Press release]. Madrid: Gobierno de España.

• OECD. (2025). Education at a glance 2025: OECD indicators. Paris: OECD Publishing.