Apprenticeship and youth unemployment | VOX, CEPR Policy Portal
It is widely believed that apprenticeships lead to better employment outcomes. This column presents the results of a field experiment conducted in France to show that apprentices do not perform significantly better than vocational students when they look for jobs outside the firm in which they trained. This means that the positive effects of apprenticeship on youth employment come from the retention of apprentices in their training firms. If the effectiveness of apprenticeship is the creation of better matches between labor market entrants and jobs, policies should focus more on the collaboration between schools and public employment services.
Empirical studies often document that apprentices have better access to employment than other vocational students after leaving school (Wolter and Ryan 2011). These facts, together with the widely publicized success of the German apprenticeship system, motivate many public policies aiming at boosting apprenticeship to foster youth employment (Kuczerat 2017).
However, little is known about the reasons why apprentices may perform better at the start of their career. Apprenticeship is generally more developed in occupations and areas whose labor market is tight, making it difficult to disentangle the effects of potential specific skills of apprentices from the demand of firms for these occupations. Potential selection of individuals with specific abilities into apprenticeship implies that estimating the impact of apprenticeship on access to jobs is difficult. Furthermore, the higher employment rate of apprentices may be the consequence of retention in their training firm, without providing any advantage in access to jobs in other firms. Thus, to know whether and how apprenticeship really fosters the integration of youths into employment, it is important to answer the following question: How do employers compare identical graduates of the same diploma acquired either after apprenticeship or after vocational education in school?
To answer this question, we measured the chances of getting a callback from employers for unemployed youth who were formerly either apprentices or vocational students (Cahuc and Hervelin 2020). The method involves sending résumés of unemployed young applicants to actual job offers, whereby the applicants are similar in all ways except for the pathway through which they got their secondary school diploma. This strategy ensures that résumés can vary in one dimension only, which serves to identify the effects of different education pathways on the probability of callback, and consequently the preferences of employers for these pathways.
We sent 3,110 applications from January to July 2018 to job offers posted in France for cook and bricklayer positions. At the aggregate level, we detect no difference in the callback probability of apprentices and vocational students. This result holds true for both occupations. It also holds true for small and large firms and for temporary and permanent jobs. The only small difference, to the advantage of apprentices, arises in commuting zones where the unemployment rate is high. This is consistent with a situation in which employers have a slight preference for apprentices, which has an impact on callback probabilities only if employers can choose among a large pool of applicants.
Relying on the Génération survey, which provides a large representative sample of students leaving education, we generate descriptive statistics showing that the findings of our correspondence study are consistent with the overall school-to-work transitions of apprentices and vocational students in France. As shown by Figure 1, on average, the unemployment rate of apprentices is 10 to 15 percentage points lower than that of their counterparts right after graduation.
Figure 1 Unemployment rate of apprentices and vocational students after upper secondary education in France
Source: Génération survey, CEREQ, 2001, 2007, 2010,2013
This gap corresponds to the difference between the share of apprentices who remain in their training firm and the share of vocational students who remain in the firm where they were interns before leaving school. Data from the Génération survey also show that, conditional on observable characteristics, apprentices do not perform better in getting jobs than vocational students once they are non-employed, whether unemployed or inactive.
It is possible that unemployed apprentices do not perform better than students in getting job offers because only the best apprentices remain in their training firm, implying that those who are looking for jobs are the less effective ones. To see whether this selection exists, we compare the wages of apprentices retained in their training firm with those of other apprentices. Their average wages are not statistically different, meaning that there is no evidence of selection of the best apprentices in their training firms. The absence of selection of best apprentices in their training firm may be explained by their propensity to quit their job at the end of their apprenticeship for personal reasons. We also find that the average wage of apprentices not retained in their training firm is not statistically different from that of students, indicating that the productivity gap between apprentices and students is at most small. This suggests that the lower unemployment rate of apprentices, compared with that of vocational students, does not originate from their higher productivity, but from retention in their training firms.
To explore this issue further, we build and estimate a search and matching model which allows us to reproduce the main stylized facts of a youth labor market with vocational students and apprentices. In this model, students and apprentices not retained in their training firm at the completion of their apprenticeship compete to get jobs. The estimation of this model shows that expanding the share of apprentices has limited impact on youth unemployment if this is not accompanied by an improvement in the retention rate of apprentices in training firms.
The conclusion that apprentices do not perform significantly better than vocational students when they look for jobs outside the firm in which they trained has important consequences for public policy. If the main advantage of apprenticeship is the creation of better matches between labor market entrants and jobs, policies should be more focused on this dimension and favor collaboration between schools and public employment services. This collaboration, which is almost non-existent in many OECD countries, is well developed in Japan and in Germany, which share important common attributes in this respect (Ryan 2001) and are very successful at integrating youths into employment. In Japan, where apprenticeship is very rare, high schools provide career support for their students (OECD 2017). Counseling and job search training are often part of senior high school curricula from the first year. In the second year of high school, many schools have specific career preparation classes for students who do not intend to pursue higher education. In the third year of high school, aspiring labor market entrants undergo a regulated job placement process at school in which the teachers responsible for career guidance match students to the available positions based on vacancy lists provided by public employment agencies. The application process follows a strict schedule to promote equal opportunities among graduates and to ensure that students focus on completing their studies. Students are not allowed to seek work independently, and employers are expected to cooperate with public employment agencies when hiring future graduates. The job placement of high school graduates is remarkably effective, averaging about 90%, and there is little evidence that it comes at the cost of lower job stability. In Germany, the Federal Employment Office recommends secondary school applicants to sponsoring employers. As in Japan, there are important interactions between schools and public employment agencies. The effectiveness of this strategy is also stressed by Noelke (2014) who argues that economic liberalization in post-socialist countries like Hungary has made the transition from vocational education to work more difficult by breaking linkages from schools to employers that performed a critical matching function.
Our findings suggest that the German-Japanese strategy targets an important cause of youth unemployment: the difficulty for job market entrants in finding jobs to which they are suited. Hence, improving the job placement of school leavers through the active involvement of public employment services in schools may be an important lever to boost youth employment.
Cahuc, P and J Herverlin (2020), “Apprenticeship and Youth Unemployment”, CEPR Discussion Paper No. 14621, April 2020.
Kuczerat, M (2017) “Incentives for Apprenticeship”, OECD Education Working Paper No 152.
Noelke, C and D Horn (2014), “Social Transformation and the Transition from Vocational Education to Work in Hungary: A Differences-in-differences Approach”, European Sociological Review 30 (4): 431–443.
OECD (2017), “Investing in Youth: Japan”, OECD Publishing, Paris.
Paul R (2001), “The School-to-Work Transition: A Cross-National Perspective”, Journal of Economic Literature 39 (1): 34–92.
Wolter, S C and P Ryan (2011), “Apprenticeship”, The Handbook of the Economics of Education 3: 521–576.