EC[ON]OMY

From school to workforce: preserving human capital

Whenever a new international ranking of school students is released, the conversation focuses on one thing: where the country placed and how many points it finished ahead of its neighbours. It sounds like the right question, but it is not. Far more important is how much of the human capital built through years of schooling and public investment survives into adulthood and becomes productivity and income, and how much simply evaporates because the economy fails to offer graduates jobs in which their knowledge is needed.

A new analysis by Francesco Avvisati, Ani Mayercord and Nathanael Reinertsen for the Organisation for Economic Co-operation and Development compares people’s test results as teenagers with their performance as adults many years later. It offers an uncomfortable but precise answer. Skills acquired at school are not fixed for life. They either continue developing through education, training and work, or fade through lack of use. Adult literacy and numeracy tend to peak at around the age of 30. What happens after that depends on whether people use those skills every day.

The strongest evidence concerns not skill levels but their trajectory. If the comparison focuses not on the scores themselves but on how countries’ results changed over time, whether school performance rose or fell, the relationship with changes in adult test results in those same countries years later is virtually zero and in some cases even negative. A country can spend years investing in better school scores and reporting progress, with almost no effect on how the same people perform as adults. Moving up the rankings does not guarantee stronger national human capital a decade later.

Another pattern makes the picture even less comfortable for enthusiasts of school rankings. The longer the interval between the test taken as a teenager and the follow-up assessment in adulthood, the weaker the relationship between the two results becomes. Among those retested at 20, the connection with their school scores remains fairly strong. Among those reassessed at 26, it is noticeably weaker. A school score is therefore less a lifelong diagnosis than a starting point whose significance fades with every year of adult life spent without using those skills.

The next question is what determines whether this capital is preserved or loses value. The answer lies not in school but in what happens immediately afterwards. Where people leave school for high-quality pathways genuinely connected to the labour market, such as relevant further education, internships and work in their field, skills do not merely survive but continue to grow. Where those pathways are fragmented, detached from employers’ actual needs or simply absent, erosion begins quickly, particularly among those who leave education early or enter informal employment.

It is telling that the same model can produce different results in different countries. Germany and Denmark, both of which have well-developed dual vocational education systems, show stable or improving skills among young adults. Austria, whose system looks much the same on paper, does not show the same pattern. Italy, Korea, Slovakia and Spain, where the transition from education to work is weaker, show flatter or declining trajectories. What matters, then, is not whether a model formally exists but whether it genuinely works: whether a qualification or internship actually opens the way to a job in which the acquired knowledge is used every day rather than left to gather dust.

Country comparisons offer a direct illustration. Adults in Finland, Norway and Slovakia perform better than their teenage scores would have predicted. Human capital in these countries is not merely preserved but continues to develop. Korea, Poland, the United States and New Zealand show the opposite effect. The gap between what students achieved at school and how those graduates performed as adults suggests that some of the capital accumulated during their education was lost. The authors make an important qualification, however. Part of the gap may reflect differences in measurement rather than the loss of knowledge. For Poland, for example, they note an unusually high share of adult respondents showing signs of low engagement with the assessment, which may have depressed the final result regardless of their actual skills.

That qualification is not incidental. It points to a broader problem that the authors examine separately. Teenagers take their test at school, in a classroom, under a teacher’s supervision and within a fixed period. Some tasks involve extended answers assessed by a human marker. Adults take their test at home, one-to-one with an interviewer, without a time limit and without open-ended questions requiring a written response. These differences are not harmless. Weaker respondents are far more sensitive to format, fatigue and motivation than stronger ones, while high performers tend to produce stable results under almost any conditions. Part of the observed gap between school and adult scores is therefore not a pure loss of human capital but a side effect of how that capital is measured. This should be considered before any country is given a definitive diagnosis.

The most uncomfortable part of the central finding is that any genuine loss of human capital is distributed unevenly. A separate longitudinal study by Vicky Jakobsen of the Danish National Centre for Social Research, cited in the paper, followed the same people from a teenage assessment to another test at the age of 30. Most of those who were at the bottom of the distribution at school remained there as adults, but one in ten managed to rise into the strongest group. The middle of the distribution was the most mobile: almost one-third moved down over time, while nearly as many moved up. Women, children of immigrants and people without higher education were notably more likely to fall from a stronger position. The loss of capital is concentrated among groups with weaker ties to the labour market rather than being spread evenly across the population.

The value of this finding is that it connects school statistics directly to money. Adult skills have long been reliable predictors of employment, earnings and individual productivity. The debate over converting school knowledge into adult human capital is therefore ultimately about who will earn more in ten years and whose qualifications will remain an unrealised line on a CV. It is telling that Avvisati, Mayercord and Reinertsen themselves consider the current link between the two assessments too weak to support precise conclusions. By the start of the next decade, they propose building a more direct statistical bridge between school and adult tests through common tasks and longitudinal tracking of the same people. If even the architects of the methodology need a more precise instrument to answer the question, governments certainly should not take comfort from a school ranking alone.

For Kazakhstan, which invests substantial public funds in schools and universities and closely follows its position in international rankings, this is a reason to change the question. What matters is not how many points 15-year-olds score in the next testing cycle, but whether the economy meets them several years later with jobs in which their knowledge is genuinely needed and valued, or whether they end up in roles where those skills have no use and gradually disappear. Until the country develops its own way of tracking what happens to graduates’ skills five and ten years after they receive their diplomas, a significant share of education investment will continue to lose value invisibly. It may leave no trace in official reports, but it will show up clearly in labour productivity and payrolls.

The conclusion is simple, which makes it more troubling. A school score is a snapshot, not a guarantee of future competence. The real return on education is measured not by a place in a league table of 15-year-olds, but by how much of that result remains after ten years of working life. It is a far more difficult but much more honest question than the familiar race up the rankings.

Elvira Baturova, EconomyKZ Research Group, exclusively for EconomyKZ.org

Scroll to Top