In response to recent technological changes and the worsening outcomes of non-college-educated workers (Autor 2019), governments around the world are becoming more interested in whether different types of secondary education (vocational versus general) might play a role in providing young people the skills they need to succeed after they graduate (European Commission 2010, US Department of Education 2013, 2018). In stark contrast to the growing body of evidence on the impact of various fields of study in higher education (Altonji et al. 2012, Hastings et al. 2013, Kirkeboen et al. 2016), there exists a paucity of compelling causal evidence on the impact of secondary-school curricula on labour-market outcomes (Altonji et al. 2011, Hampf and Woessman 2017, Hanushek et al. 2011, 2017).
Understanding the potential consequences of secondary-school curricula is particularly important given that this choice takes place before higher education and, for many people, is the highest level of education before entry into the labour market. Further, the availability of vocational secondary education is one of the largest differences among national education systems (Figure 1).
Figure 1 Enrolment in vocational and general secondary education in OECD countries
A common view suggests that there may be a trade-off between benefits of vocational education in the short term and adverse impacts later on (Krueger and Kumar 2004, Hampf and Woessman 2017, Hanushek et al. 2017). According to this literature, vocational education may provide applicants with occupation-specific skills that better facilitate the initial school-to-work transition. Further, vocational education may offer an important alternative for youth otherwise at risk of dropping out of secondary education.
On the other hand, general education has been thought to better prepare applicants for further education – thus enhancing labour-market prospects later in the career. Moreover, with changes in technology and the future of work, critics fear that vocational skills may become obsolete at a faster rate than general skills.
Such trade-offs are in line with the trends in mean outcomes in the Finnish data on the universe of students graduating from compulsory education in Finland during 1996–2000 (Figure 2). On average, applicants admitted to the vocational track experience an initial advantage in the labour market but are overtaken by their general-track peers 11–12 years after admission (ages 27–28).
Figure 2 Mean employment (a) and earnings (b), by secondary school track
(a) Employment
(b) Earnings
Seventeen years after admission to secondary education (age 33), applicants admitted to the vocational track annually earn €4,000 less than applicants admitted to the general track and are employed fewer months per year. Of course, these mean differences may be driven by selection.
Overcoming selection using admissions cut-offs from Finland
To examine the labour-market returns to vocational versus general secondary education, we use a regression discontinuity design created by the centralised admissions process in Finland (Silliman and Virtanen 2021). In this setup, we focus on applicants to secondary education who apply to both vocational and general tracks and whose path is determined by randomness in admissions cut-offs to oversubscribed schools (Figure 3).
Figure 3 Admissions (a) and (b) enrolment across the cut-off
(a) Admissions
(b) Enrolment across the cut-off
Prior empirical work aiming to identify the causal effect of vocational secondary education provides evidence that vocational education can improve short-term outcomes. Recent papers exploiting randomness in admissions to oversubscribed US schools from Massachusetts, Connecticut, and North Carolina in the US suggest that vocational education can improve on-time graduation but may have mixed effects on enrolment in higher education (Dougherty 2018, Hemelt et al. 2019, Brunner et al. 2019).
Further, evidence from a randomised control trial targeting disadvantaged communities in the US suggests that increasing the vocational component of secondary education boosts earnings after graduation (Kemple and Willner 2008).
However, comparing the labour-market outcomes of graduates from vocational and general programmes across European countries over their life cycles, researchers argue that the benefits of vocational education may be short-lived, with the initial annual wage premium of vocational education disappearing by the early thirties (Brunello and Rocco 2017, Hanushek et al. 2017, Hampf and Woessman 2017).
In contrast, a second approach to exploring the longer-term effects of vocational secondary education has focused on national reforms and finds no benefits of increased exposure to general education. A study of a reform in Romania that shifted a large proportion of students from vocational training to general education suggests that the benefits of general education are largely driven by selection (Malamud and Pop-Eleches 2010, 2011).
Studies in the Netherlands and Sweden which look at vocational-education reforms that increased the general content in the vocational track find no benefits of additional general content on labour-market outcomes (Oosterbeek and Webbink 2007, Hall 2016). In Norway, Bertrand et al. (2019) find that a similar reform also increased selection into the vocational track and thereby led to improved earnings for those induced into vocational education.
Our findings from Finland provide local estimates for individuals most likely to be impacted by changes in the size of the vocational-education sector. Moreover, while other research has relied primarily on reforms that affect the educational choices of entire cohorts or cross-national differences in secondary sectors, the regression discontinuity design allows for cleaner inference by comparing individuals within the same age cohort and working within the same labour market. As observed by Bertrand et al. (2019), effects estimated using vocational-education reforms can be driven by compositional changes related to track choice across cohorts as well as changes in the content of the vocational track.
Access to the vocational track can boost income over the long term
Causal estimates from our study suggest that enrolment in vocational secondary education increases initial annual income – and this benefit persists through age 33 (a 6% boost 17 years later), with no effect on months of employment, for applicants at the margin of admission to vocational versus general education (Figures 4a and 4b).
Figure 4 The effect of admission to vocational education on (a) earnings and (b) employment
(a) On earnings
(b) On employment
Further suggesting that the returns may not turn negative over the even-longer term, admission to the vocational track does not reduce the likelihood of ever graduating from higher education, and graduates are no more likely to be employed in occupations at risk of automation or offshoring.
Results from present discount value calculations of the lifetime return to vocational education under several scenarios suggest that it is highly unlikely that the lifetime vocational premium will turn negative through retirement. The effect is heightened for applicants who indicate a preference