Credential and Skill Mismatches Among Tertiary Graduates: The effect of labour market institutions on the differences between fields of study in 18 countries

ABSTRACT

This study provides new empirical evidence about tertiary graduates’ overeducation, by analysing the influence of labour market institutions on the incidence and distribution of the phenomenon across fields of study. In particular, the analyses focus on the level of employment protection, the regulation of access to the so-called liberal professions, and the propensity of welfare states to hire skilled workers. Data from two comparative surveys – REFLEX and HEGESCO – are used, and a wide set of information is employed to split overeducation in two forms of suboptimal allocation of individuals in the labour market: credential and skill mismatches. The first term refers to the mismatch between formal educational credentials and job requirements, whereas the second term refers to the mismatch between the skills acquired through education and those needed to perform a job. Results suggest that field of study differentials vary by country and that welfare and labour market institutions illuminate these cross-national variations. Moreover, the results support the claim that it is necessary to distinguish between credential and skill mismatches, showing that these institutional factors do not always affect them similarly.

Key words:

• overeducation
• skill mismatch
• field of study
• labour market institutions

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 For a more detailed discussion, see McGuinness (2006) and Leuven and Oosterbeek (2011).

2 Frequentist methods find maximum likelihood point estimates for the parameters of interest in the model by iterating between two deterministic steps until two consecutive estimates for each parameter are sufficiently close together, thus achieving convergence. Conversely, MCMC methods are simulation-based procedures, which are run for many iterations, each of which produces an estimate for each unknown parameter. These estimates are not independent because, for each iteration, the estimate for the last iteration is used to produce the next estimate. Thus, these methods produce accurate interval estimates, from which it is possible to calculate the posterior mean and standard deviation (Browne 2003).

3 An additional factor that may affect the incidence and distribution of both credential and skill mismatches, by balancing the supply of professionals that compete for the vacant positions, is the openness of the corresponding educational system. Unfortunately, comparable data on this topic are not available.

4 Unfortunately, we can use only a general indicator of professional services regulation, which does not distinguish among different professional sectors. Further research could use more detailed, field-specific indicators.

5 The role of educational institutions is not a major focus of this study, particularly because several previous studies have documented it extensively. However, the finding that age of tracking does not matter does not contradict these studies once we consider that the models presented here control for graduation rates. In other words, the effect of stratification is entirely explained by this variable, suggesting that early tracking affects overeducation mainly by restricting the supply of graduates.

6 Some robustness checks have been conducted by excluding the country-level control variables and adding the interaction terms one by one. The results do not substantively change (the results of these models are available upon request to the author).

Additional information

Giulia Assirelli is a Ph.D. candidate in Sociology and Social Research at the University of Trento, Italy. Her research interests include transitions from education to labour market, comparative sociology, and quantitative methods of data analysis.