Patterns of Overeducation in Europe: The Role of Field of Study

In general, the term overeducation refers to a job match in which the educational level of the worker clearly exceeds the educational requirements of the job. In the terminology of labor economics, this is often considered a vertical skill mismatch, as opposed to horizontal mismatches (workers choosing jobs with requirements outside the scope of their field of study/apprenticeship). A widespread occurrence of this phenomenon can seriously impair the competitiveness of an economy. From a macroeconomic perspective, an overeducation status of qualified workers reflects a waste of scarce human capital. From a microeconomic perspective, it can affect a worker’s job satisfaction. In turn, a skill mismatch can reduce overall work motivation, expressing itself in more frequent absenteeism and higher turnover of the workforce (Tsang and Levin, 1985; Sicherman, 1991; Sloane et al., 1999). Moreover, overeducation is associated with earnings losses (Daly et al., 2000; Bauer, 2002; Boll and Leppin, 2016).

However, before being able to tackle the problem successfully, it is essential to understand the driving forces of overeducation at the individual level. There is a vast body of literature including meta-studies documenting the importance of job-related and micro-level factors as well as country specificities in the context of overeducation. However, most investigations address either one or only a few countries in the same methodological and data setting, which hampers a substantial cross-country comparison. Moreover, few studies investigate the role of field of study, although from a theoretical point of view, the magnitude of skill mismatch should relate to structural features of field-specific labor market segments. Finally, many studies lack a substantial sensitivity analysis, although it could be shown that the choice of the educational standard heavily impacts the results (Battu et al., 2000; Verhaest and Omey, 2006; Quintini, 2011; Boll et al., 2016).

Against this background, the aim of this paper is to identify possible determinants of overeducation for young (20–35 years) highly educated (tertiary level) workers in EU-28 countries with a special focus on the role of field of study. We make use of the 2016 wave of the European Labour Force Survey (EU-LFS), a quarterly household sample survey that covers approximately 1.8 million individuals aged 15 years or older. We contribute to the literature in three ways. First, we investigate the role of field of study as a so far widely neglected factor in the context of overeducation. We are able to identify the significance of this factor by controlling for a range of alternative explanatory factors. Second, we present country-specific results for 21 EU countries together with a cross-country estimation within the same methodological framework. This helps to identify differences in the seriousness of the phenomenon between countries and to develop tailor-made policy recipes. Third, we provide a solid robustness check of our results, deploying two more measures of overeducation which largely confirm the results of our benchmark measure.

Our findings reveal different overeducation risks for graduates from different fields. Compared to social sciences, male graduates from, for example, Education, Health and Welfare, Engineering, and ICT (Information and Communication Technologies) are less and those from, for example, Services and Natural Sciences are more at risk. These findings hold for the majority of countries and are robust against a change of the standard education. However, some fields show gender-specific risks. We suggest that different degrees of job-specificity, productivity signals, and gender stereotypes answer for the cross-field and cross-gender differentials. Moreover, country fixed effects point to relevant structural differences between national labor markets and educational systems.

The empirical literature on the topic of overeducation has come up with a wide range of findings on the influence of some individual- and job-related factors, in particular work experience (Alba-Ramirez, 1993; Groot, 1996; Sloane et al., 1999; Nielsen, 2011, Boll et al., 2016) and job tenure (Büchel and van Ham, 2003; Büchel and Battu, 2003; Groot and van den Brink, 2003; Ortiz, 2010), beyond macro-level factors like job scarcity on the labor market, which might advantage graduates due to lower expected training requirements on the side of employers (Thurow, 1975).

Much less well-documented is the role of the educational field of a person, which constitutes an important element for predicting labor market outcomes (van de Werfhorst and Kraaykamp, 2001; Hansen, 2001) and hence might also be a determinant of overeducation. Several reasons motivate this view. First, fields of study differ in their occupational focus. Fields like medicine or engineering with their quite narrowly defined job profiles might require more occupation-specific skills, raising the chances of graduates to find appropriate jobs in the corresponding occupational groups (Reimer et al., 2008). The high job-specificity protects graduates of fields like medicine, law, or architecture from educational mismatch (Ortiz and Kucel, 2008). Second, credentialism theories suggest that in a world where the true personal abilities are unknown, the chosen field of study can also act as an ability signal to employers. Obtaining a degree in fields like maths, natural sciences, or technical disciplines, which enjoy the reputation of imposing high intellectual demands on their students, could convince employers of the extraordinary talent and/or motivation of applicants (Barone and Ortiz, 2011). This could give them preferred access to positions with high skill requirements, possibly also outside the occupational groups associated with their subjects. Third, field choice might be triggered by individual gender role orientations and social origin (Polachek, 1978; Bradley, 2000), such that field-specific labor market outcomes are not purely causal effects but to some part driven by selection into fields. More specifically, gender norms might impact decisions on family formation and marriage and via this channel impact educational choices (Chiappori et al., 2009; Attanasio and Kaufmann, 2017). Beyond horizontal segregation, also vertical segregation relates to gender norms. Whereas women still bear the lion’s share of household chores, they are underrepresented in managerial positions throughout Europe. For example in April 2015, the EU-average of women’s share among senior executives of the largest nationally registered companies listed on the national stock exchange amounted to 13.7% and their share among nonexecutive directors stood at 22.5% (European Union, 2016, p. 26). Both the choice of “female” occupations at the beginning of the career and the typically “female” decisions in its subsequent stages can be associated with the underutilization of formal education in the current job. Furthermore, having graduated as a female in a male dominated field could convey a negative productivity signal to employers, relative to male graduates in the same field, resulting in ceteris paribus higher overeducation.

The impact of the chosen field on the risk of overeducation is also likely to differ with educational level. The training received by graduates from tertiary education is typically of a more academic nature and less focused on occupation-specific skills than vocational programs. In the first analysis of this kind, Green and McIntosh (2007) restrict their estimation for the United Kingdom to the subsample of university graduates. Also some later studies focused on tertiary graduates only (Reimer et al., 2008; Smyth and Steinmetz, 2008; Tarvid, 2012; Berlingieri and Zierahn, 2014; Capsada-Munsech, 2015).

The comparability of results is limited by the different methodological settings that have been used in the literature so far, for example, with respect to field classifications, measures of educational standards, or the list of covariates. For example, in the analysis of Green and McIntosh (2007), who make a quite detailed distinction between 12 educational fields, degrees in Physical sciences and in Computing are estimated to lower the overeducation probability significantly relative to the reference category Business and Management Studies.

The insignificance of the field Math explain the authors by the fact that school grades in Maths were included as an additional control variable, thereby diluting the measurement of the field effect.

Also Tarvid (2012) who makes use of the European Social Survey data and tests the field effect in a supranational sample comprising 30 countries finds that graduates from Services exhibit a much higher overeducation probability than graduates from business, law, and economics. Probabilities lower than for the reference were detected for the fields education and health. The results of Berlingieri and Zierahn (2014) for German male graduates support the notion of a low risk for graduates of natural sciences, compared to business and law. Finally, the most recent test we are aware of was conducted by Capsada-Munsech (2015) for Italian university graduates. The study reports the lowest overeducation probability for Medicine, compared with the reference category humanities.

To sum up, the literature so far suggests that, albeit a considerable degree of heterogeneity in methodology and data, students of Social sciences, Services and Humanities tend to be at higher risk than those in Natural sciences and ICT. This supports the notion that there is a linkage between occupational closure and the risk of overeducation. Highly job-specific programs like medicine seem to represent generally acknowledged entry requirements for the corresponding occupations, which restrict job competition to the graduates from these fields.

With respect to gender-specific field-of-study effects one could think of six (nonexclusive) channels. First, the risk discrepancies might reflect that, on average, female graduates exhibit different preferences in terms of job attributes than their male counterparts in the same educational fields (Coudin et al., 2018). Second, they might be a sign of field-specific gender discrimination concerning access to adequate jobs, for example, as a consequence of gender stereotypes regarding job images (Glick et al., 1995). Third, they could indicate that in these fields male graduates (for Services and Natural sciences) and female graduates (for Arts and humanities) showed on average the better academic performance, giving them better chances to enter adequate positions. Fourth, they could also indicate the existence of educational sorting at a lower aggregation level than measured. Fifth, gender differences in field-specific risks could origin in masked gender differences regarding assumed occupations.

Note that although we specify overeducation as an occupation-specific risk (see Section 3), this does obviously not prevent genders from sorting into different occupations within the same educational field.

Contrary to fields where women are underrepresented (ICT, Natural Sciences, and Engineering) and fields where women dominate among university leavers (education, health and welfare, art and humanities, and social sciences), women and men are almost equally distributed among university leavers in the field of Services (OECD, 2016). Combined with a lower employment rate of women (also) in this field compared to their male counterparts, this could lead to a relative scarcity of highly educated women in Services-related occupations on the labor market. Contrary to STEM professions where women often suffer negative productivity signals arising from gender stereotypes, there is hardly any comparable stigma in service-related professions. In sum, this might trigger a lower overedcuation risk of female graduates in the field of services.

The study at hand aims to verify the linkage between job-specificity and overeducation in a consistent methodological and data framework. In particular, we hypothesize that those study programs that train graduates for a more clearly defined range of occupations are associated with a lower overeducation risk than less job-specific study programs. Further, based on the above discussed channels of gendered preferences, gendered ability signals and demand/supply ratios on the labor market we expect a comparatively lower overeducation risk of female graduates in the field of services compared to their male counterparts. In order to validate the sensitivity of our results to the measurement methods, we compare our baseline measure of overeducation with two alternatives.

We use data from the European Labour Force Survey (EU-LFS)

For more detailed information on the European Labour Force Survey, see, for example, European Union (2014).

In line with previous studies (Reimer et al., 2008; Smyth and Steinmetz, 2008), we restrict our sample to highly educated individuals, as the issue of overeducation is by definition most relevant for members of this group and, with a sharp increase of graduates’ population shares during the last decades in OECD countries (from 23.3% in 1995 to 43.1% in 2016 on average), affecting more and more people (OECD, 2018).

Studies that compare educational groups stress the higher magnitude of overeducation for graduates. For Germany for example, a study based on the Socio-economic Panel (SOEP) estimated a 30% (41%) risk for West (East) German male graduates and a 36% (38%) risk for West (East) German female graduates of statistical overeducation in 2011, whereas the corresponding figures for workers with medium education are 8% (12%) for men and 14% (10%) for women (Boll et al., 2016).

We refer to the above-mentioned overeducation as a vertical inadequacy. In the literature, different ways for measuring overeducation are followed, from expert evaluation of occupation-specific required education (which is seldomly available, Eckaus, 1964) and respondents’ subjective assessments to statistical approaches (realized matches). Results often change when subjective evaluations of overeducation are used instead of the statistical measure (Bauer, 2002; Chiswick and Miller, 2010; Nielsen, 2011; Boll et al., 2016). For our purposes, we adopt the variant of the realized matches approach. This is the only measure that can be employed based on the data at hand, but referring to the literature, each measure has its pros and cons. Empirical evidence suggests that self-assessed overeducation is subject to other job features such as occupational status and particularly income (Dolton and Vignoles, 2000). Survey participants may be inclined to exaggerate educational requirements of their job for various reasons (Borghans and de Grip, 2000). Furthermore, self-assessed overeducation might be gender biased (Leuven and Oosterbeek, 2011).

More specifically, we follow the realized matches approach proposed by Kiker et al. (1997) and code a person as being overeducated if his or her highest educational attainment level is higher than the benchmark education level of her occupation group at the two-digit ISCO level.

The mean value is not an available option since the data provides us with educational categories only. Hence, the main advantage of the mode over the mean that it is less sensitive to outliers (Kiker et al., 1997) does not apply in our data context.

The choice of reference point can potentially have a sensitive impact on the measurement, depending on the specific distributions of educational levels within an occupation group. Preferring the 80th percentile over the mode follows the idea that the mode regularly relates to higher overeducation rates in the same methodological setting and based on the same data (see for a literature overview Cedefop, 2010, pp. 18–20). This particularly applies when the underlying distribution of the dependent variable is fairly even; in this case, depending on the exact position of the most frequent single value the observations above (or below) this threshold may cover a quite high population share.

With the “ISCO measure,” we define those graduates to be overeducated who exhibit an ISCO one-digit level of 4 or higher (ISCO-08 = International Standard Classification of Occupations, cf. ILO, 2016). In other words, individuals working as managers (ISCO-08 level 1), professionals (ISCO-08 level 2), technicians, and associate professionals (ISCO-08 level 3) are considered not to be overeducated. The assumption underlying the ISCO measure is that demanding jobs require a certain educational level. Note, however, that the ISCED levels define the skill requirements of major occupational groups. Particularly, ISCO-08 level 3 (2) corresponds to ISCED-2011 level 5 (6, 7, and 8). It becomes evident that the ISCO measure deviates from the realized matches approach in that it defines educational mismatch from theoretical reasoning whereas the 80th percentile and the modal value both rely on the realized statistical distribution of education among employees within an occupation. Principally, the median would have been another default option, but as a 50th percentile criterion, it is technically similar to our 80th percentile benchmark measure. The only difference is that it sets a lower hurdle, which unsurprisingly leads to very high and rather unplausible overeducation rates. Therefore, we refrain from implementing it.

We implement the field of study indicator provided in the EU-LFS data as an explanatory variable. Following the classification scheme ISCED 2013-F, it distinguishes 11 field categories. In our estimation model, the single categories are coded as categorical dummy variables, choosing the category “Social Sciences, Journalism and Information” to be the omitted reference category. Additionally, in order to illuminate potential gender differences in the role of the single fields, we include interaction terms of the field category with a dummy measuring sex (female = 1; male = 0).

The associations of fields of study are isolated by controlling for personal and job characteristics. Personal characteristics include sex and foreign nationality. Furthermore, we consider the impact of age within our already narrow sample of 20–34 years old by introducing dummies referring to the 25–29 years old and the 30–34 years old, respectively, with the 20–24 years as a reference.

Among job characteristics, usual working hours are given as the number of hours that a respondent is usually working per week in her main job. Tenure is defined as the number of years since a person started to work with her current employer or as self-employed. In order to shed light on potential nonlinearities, tenure is also included as squared term. Further, we include a temporary contract dummy. Firm size is split into three dummy variables, namely 11–19, 20–49, and more than 50 employees, respectively. Smallest firms with 10 employees or less serve as a reference. Finally, we include 20 sector dummy’s (sections according to NACE Rev. 2) and country dummies in our regressions.

In a preceding Heckman correction for employment selection (see Estimation method), we use a set of household characteristics which are deemed to exert an influence on employment propensity but not on overeducation probability. The household characteristics compose of the spouse’s educational level (ISCED) and dummys for unemployed or inactive adults. Based on assortative mating (Mare, 1991), workers living together with unemployed might on average be less productive themselves, a fact that reduces their chances to find a job. Further, we account for elderly persons (aged 75 and older) and children by age of the youngest one (0–5, 6–11, and 12–17 years) in the same household. As recent research still identifies a strong gender bias in the labor market implications of children (Waldfogel, 1998), interaction terms with sex are included for this last indicator.

The most simple (and also most common) approach to analyze impact factors on overeducation risk is to implement a Probit model (see Judge et al., 1988). The target variable y_i classifies a respondent either to be overeducated (y_i = 1) or not (y_i = 0). In the Probit model, the probability of y_i = 1 is modelled as follows:

where ϕ (.) is the cumulative distribution function of the standard normal distribution and X is the set of covariates presented above. The model can be estimated with the Maximum-Likelihood-Method, which yields consistent, asymptotically efficient and asymptotically normal distributed estimates. Due to the nonlinearity of the model, marginal effects are not simply given by the estimated coefficients β^,$\hat{\beta },$but depend on the level of the covariates. In the Results section, we report average marginal effects, that is, averages of the specific marginal effects determined for each observation. In comparison to the estimated coefficients as such, this facilitates interpretation, as the actual distributions of the explanatory factors are accounted for.

A drawback of this simple approach is that it neglects a potential estimation bias due to self-selection into employment. It rests the analysis purely on those individuals having a job at the time of observation. However, intuition suggests that overeducation risk could well be correlated with employment selection, for instance if the prospect of entering into a skill mismatch induces job seekers to rather stay unemployed to circumvent expected earnings drawbacks or other disadvantages like job dissatisfaction. Under such circumstances, employed and nonemployed individuals systematically differ in their risk levels. Results based on estimations not accounting for the impact of employment selection will then be biased in the sense that the overeducation risk of persons affiliated to high-risk fields would be underestimated. To correct for selection bias, we use the two-step correction mechanism (Heckman, 1979) where some household variables serve as identification variables (see Section 3). Specifically, we assume that the absence of further adults in the household increases the employment propensity of the person since he or she is more likely to rely on own earned income whereas this factor should not be directly related to the risk of being overeducated (cf., for example, Ghignoni and Verashchagina, 2014 for a similar approach). The opposite holds true for the presence of small children in the household (although differently motivated via an increased household productivity which lowers employment propensity).

As a first step, based on a sample of workers and nonworkers, a Probit model is specified estimating the likelihood of being in employment at the time of observation as a function of individual and household-related characteristics (as well as country dummies in the cross-country estimation), wherein the household characteristics serve as identification variables. As a second step, based on a sample restricted to workers, the Probit model for overeducation can be estimated, including the inverse mills ratio computed from the results of the first step as an additional control variable, reflecting the impact of employment selection to the overeducation equation.

The purpose of this paper was to conduct a comprehensive econometric analysis of potential determinants of overeducation among graduates in 21 EU countries in a unified framework. A special focus was set on the role of field of study for graduates’ vertical mismatch incidence. It turned out that both in the cross-country estimation and at country-level differences in overeducation risk between graduates from different fields are significant. Furthermore, gender discrepancies in the impact of certain fields are noticeable. At the European level, graduates from Services, Natural sciences, and Agriculture are found to exhibit the highest risk among men. At the same time, male graduates from fields like ICT, Health and welfare, Education, Engineering but interestingly also Arts and humanities, are exposed to a rather low risk. The field-specific risks apply for the majority of countries and are robust against measure changes regarding the educational standard. We suggest that the degree of job-specifity of study programs significantly shapes the cross-field differentials in overeducation risks.

Gender differences in field-specific overeducation risks mostly lack statistical significance, with Services and Natural sciences, where female graduates are assessed to be at significantly lower risk than male graduates and Arts and humanities, where the opposite is true marking the exceptions. The highly negatively significant association of Services supports our expectations. We suggest that gender differences in preferences, field-specific productivity signals, and demand/supply ratios on the labor market might answer for the comparatively lower overeducation risk of women in the field of Services.

Moreover, country fixed effects point to relevant structural differences between national labor markets and educational systems. As we included, a selection correction in our estimation approach, country differences concerning employment selection should not be the source of this heterogeneity. Rather, differences in educational systems, in the capacities of labor markets to absorb young tertiary graduates as well as in culture- and tradition-based attitudes seem likely candidates. Although we made some references to the literature here, disentangling these different national-aspects and utilizing them for an analysis of country patterns represents a second interesting avenue for future research.

Further arguments add to the limitations of our study. Despite the wide range of individual covariates, we are aware of missing factors that proved to be relevant for overeducation propensity like paternal background (Jackson et al., 2008) or students’ academic ability before enrolling in higher education (Barone and Ortiz, 2011). With the underlying econometric approach, causal interferences must not be drawn.

We make use of the 2016 wave of the European Labour Force Survey, a quarterly household sample survey that covers approximately 1.8 million individuals aged 15 years or older. This data set provides rich information on the respondent’s demographic background, labor status, employment characteristics, and educational attainment. It allows us to assess and compare the impact of a large variety of potential determinants of overeducation, both separately for single countries and in a cross-country estimation (21 EU countries). Our particular interest lies in the role of fields of study for vertical educational mismatch. Therefore, we restrict our sample to highly educated individuals, as the issue of overeducation is by definition most relevant for members of this group. Highly educated individuals are defined as persons who have completed tertiary education. This corresponds to educational levels 6, 7, and 8 of the ISCED 2011 classification included in the dataset. Furthermore, the sample is restricted to respondents aged 20–34 years. We refer to overeducation as a vertical inadequacy and adopt the realized matches approach Kiker et al. (1997). We code a person as being overeducated if his or her highest educational attainment level is higher than the 80th percentile of the distribution of observed levels of education in the given occupation. As sensitivity checks, we additionally report results calculated based on the mode as the educational standard as well as based on an “ISCO measure.”

In order to analyze the impact factors on overeducation risk, we estimate Probit models via maximum–likelihood method. Furthermore, in order to control for self-selection into employment, we apply the two-step Heckman approach (Heckman, 1979). As a first step, based on a sample of workers and nonworkers, a Probit model is specified estimating the likelihood of being in employment at the time of observation as a function of several individual and household context variables. As a second step, based on a sample restricted to workers, the Probit model for overeducation can be estimated, including the inverse mills ratio computed from the results of the first step as an additional control variable, correcting for the impact of employment selection in the overeducation equation.