(Mis-)matching in Germany

Workers do not always have jobs perfectly matched to their skills and abilities. Research on the question as to what constitutes a good match is usually based on the empirical correspondence between workers' formal qualifications and the formal qualification requirements of their jobs. However, there is no up-to-date and in-depth information on the incidence of skill-based mismatching in Germany so far. We use the BIBB/BAuA Employment Survey 2006 with rich information fitting the job requirement approach to provide up-to-date and in-depth figures on the incidence of formal and skill-based mismatching in Germany. Moreover, we study the extent to which mismatching varies with workers' socio-demographic and qualification characteristics as well as with job tasks.

Mismatching, i.e. the missing correspondence between the skills of a job-holder and the skill requirements of his job, might have negative consequences for the worker himself (e.g. in the form of wage penalties ⁰¹ or job dissatisfaction), for the firm he is employed in, as well as for the national economy. Studies of Germany and other economically advanced countries demonstrate that mismatching is a relevant phenomenon on labour markets (cf. MCGUINNESS 2006). Meta-analyses of international studies especially from the 1980's and 1990's considering estimates for Germany (cf. MCGUINNESS 2006; GROOT/VAN DEN BRINK 2000) indicate comparatively low rates of formal mismatching in Germany (cf. MCGUINNESS 2006, p. 388; GREEN/MCINTOSH 2000). However, there is no information on the incidence of skill-based mismatching in Germany so far. International research shows that the proportion of over-qualified workers exceeds the proportion of under-qualified workers in most countries. The extent to which this currently applies to Germany, and whether this is valid for both types of matching (formal and skills oriented) is a further issue requiring investigation. Moreover, some studies also show that younger workers, women and workers with a migration background are over-qualified more frequently, whereas male workers are less likely to be over-qualified (cf. MCGUINNESS 2006, p. 388; GREEN/MCINTOSH 2000).

An analysis of formal matching in Germany (POLLMANN-SCHULT/MAYER 2004) shows differences between cohorts within matching categories between various vocational qualifications below the tertiary-level. A series of studies (including BIERSACK et al. 2008; FEHSE/KERST 2007) addresses the issue of adequate employment of tertiary-level graduates from various disciplines. The Educational Reporting Consortium (2006, pp. 185 ff.) has shown that formal over-qualification has increased for graduates from academic tertiary-level institutions from 1984 and 1995 to 2004 (19.8%) and, currently, is slightly above the level of over-qualification of graduates from (upper secondary) vocational education and training institutions (2004: 17.2%). As far as we know, there is no study focusing on the differences between various vocational qualifications at the secondary and tertiary-level with regard to both types of matching in Germany.

Finally, some studies see signs of an increase in over-education in Europe in recent years (cf. GREEN 2006; KORPI/TAHLIN 2009). Even though we are unable to demonstrate whether mismatching has increased or decreased in Germany, to look for trends we extend our analyses by considering job tasks. We use a classification of occupational activities into cognitive (analytical and interactive) and manual routine and non-routine job tasks which are linked with assumptions regarding their labour market demand over time (cf. AUTOR et al. 2003). Thus, whereas the demand for cognitive and manual routine job tasks should decline, the demand for high-skilled cognitive and (simple) manual non-routine activities should increase over time (cf. Table 3, see below; for evidence of such a development in Europe, cf. GOOS/MANNING/SALOMONS 2009).

A common feature of studies on mismatching in Germany is their focus on formal matching (DALY/BÜCHEL/DUNCAN 2000; BÜCHEL 2002; BAUER 2002).

The data of the BIBB/BAuA Employment Survey 2006 which was conducted by the Federal Institute for Vocational Education and Training (BIBB) and the Federal Institute for Occupational Safety and Health (BAuA) - see box - makes it possible to directly compare formal qualification and formal requirements for employees. In addition to this, the survey also allows for analysing the relationship between workers' skills and the requirements of the workplace.

• Federal Institute for Vocational Education and Training (BIBB) and the Federal Institute for Occupational Safety and Health (BAuA) Employment Survey 2006
  The BIBB/BAuA Employment Survey 2006 is a representative labour force cross-section on qualification and working conditions in Germany. It includes information on respondent's qualification and career history (school education, initial and continuing vocational education and training, career development and change of occupation, usefulness of vocational qualifications etc.), as well as on detailed job-related information (organizational information, job tasks, job skill requirements, working conditions, health etc.). With a total sample size of 20,000 it is directed to study special social groups (such as old-age, female, non-formally qualified workers, workers with different national backgrounds) and developments within detailed occupations, industries and vocational fields.
• Information on the BIBB/BAuA Employement Survey 2006 is available at http://www.bibb.de/arbeit-im-wandel(German-speaking). For English-speaking information on data and access to data, please visit https://www.bibb.de/en/53.php.

Incidence of mismatching

To measure formal (mis)matching we use the information on respondents' educational attainment and respondents' assessment of the typical vocational qualification that is required to do their current job ⁰² Comparing both variables ⁰³ it can be shown that approximately ten percent of workers are under-qualified and fewer than twenty percent are over-qualified in their current job. Thus, around 70 percent are matched in terms of formal qualifications (cf. Table 1). Additionally, the BIBB/BAuA survey asked workers whether, in their job, they generally feel being up to, overstrained or undertrained by the requirements against their skills. On the basis of this definition of matching, i.e. the matching between the skills and knowledge of the job holder and the job's skill requirements, more than 80 percent of workers are adequately employed (cf. ibid). As with formal matching, it is observable that higher proportions of workers are over- than under-qualified. This corresponds with results from other countries (cf. GREEN/MCINTOSH 2007; MCGUINNESS 2006). With minor exceptions (less than 5%), German workers thus feel being matched for the requirements of their jobs.

 

Table 1: Formal and skill-based (mis)matching

Note: differences due to missing values.

Source: BIBB/BAuA Employment Survey 2005/2006, weighted values, own calculations.

Table 2: Formal and skill-based (mis)matching combined

Source: BIBB/BAuA Employment Survey 2005/2006, weighted values, own calculations.

Combining the various matching measures (cf. Table 2) reveals that, to a large extent, the different types of mismatching are independent of one another. "Two-fold" over-qualification and under-qualification, i.e. an over-qualification or under-qualification in both formal and skills-based terms, are extremely rare in Germany (4.3% and 0.7% respectively). Moreover, a larger share of workers is (only) inappropriately employed in terms of their formal qualifications (13.2% and 9.9%). Skill-based mismatching without formal mismatching occurs less frequently (8.2% and 4.0%). These results indicate that over-qualification and under-qualification in Germany are comparatively low, in particular as compared to the Anglo-Saxon countries (cf. MCGUINNESS 2006; GROOT/VAN DEN BRINK 2000).

SOCIO-DEMOGRAPHIC DIFFERENCES in (MIS)MATCHING

Both formal and skill-based mismatching vary significantly with workers socio-demographic characteristics. Whereas the chance for being matched is equally distributed between men and women, the direction of mismatching varies by workers sex. The incidence of formal and twofold under-qualification is higher for men and that of formal and twofold over-qualificationis higher for women. Also, the incidence of mismatching (most notably twofold and skill-based) is higher for workers with a migration background. However, a migration background does not have an independent effect on mismatching - the bivariate tabulation hides the impact of intervening variables such as qualification, age and industry (ROHRBACH-SCHMIDT/TIEMANN 2010). An analysis of mismatching by age-cohorts shows that compared to the main working age population young workers aged 15 to 24 and old-age workers (age 65 and above) are matched less often. Whereas young workers entering the labour market may possibly accept a mismatched employment with the aim of obtaining an initial "foothold in the market", mismatched employment of workers aged 65 and over possibly is on grounds of securing any kind of earning opportunity.

QUALIFICATION RELATED DIFFERENCES in (MIS)MATCHING

Relating mismatching to different qualification levels (highest level of vocational education attained) reveals some decisive matching patterns (cf. Figure): Firstly, workers with an apprenticeship training certificate (in the German vocational education and training (VET) sector) and university graduates perform equally well in getting a matched job. This clearly underlines the particular position of the vocational education and training system in Germany. In the case of over- or under-qualification in terms of formal credentials, workers with an apprenticeship do slightly better than graduates from tertiary-level academic institutions (19.4% as opposed to 17.4%). With regard to skills and knowledge, however, workers who have completed dual and school-based VET are more likely to be over-qualified than university graduates. Compared to these two groups, those without qualifications and workers who have undergone advanced training to obtain a master craftsman or technician qualification are significantly less likely to have a matched job. However, the completion of a master craftsman or technician qualification mostly leads to over-qualified employment only from a formal point of view, not necessarily with regard to the required skills.

Figure: (Mis)matching by qualifications (in percent)

JOB TASK RELATED DIFFERENCES in (MIS)MATCHING

Differences within the matching categories also emerge with regard to job task related characteristics, i.e. groups of occupational activities introduced by AUTOR et al. (2003) - cf. Tables 3 and 4. Roughly speaking, cognitive non-routine tasks which exhibit a particularly rising labour market demand are less likely to be executed by formal and skill-based mismatched workers. On the contrary, mismatched workers perform cognitive and also manual routine tasks more frequently. Non-routine manual tasks are more frequent with under-qualified workers, and less frequent with over-qualified workers. Overall, the results might be interpreted as a sign for (rising) mismatching in routine tasks and in non-routine manual tasks through displacement processes.

This interpretation is supported by an analysis of occupational groups and industries (results are available on request). Lower occupational groups (ISCO-88 Main Groups 8 and 9) and commercial occupations are particularly characterised by formal and skill-based over-qualification. From all industry branches, commercial occupations also display the highest values for routine cognitive task. Against that background, a decrease in the demand for routine tasks could lead to an increase in mismatched employment in Germany as in other European countries.

Table 3: Operationalisation of occupational activities within the routine/non-routine scheme developed by AUTOR et al. (2003)

Task group	Typical level of qualification	Assumed labour market demand	Item in BIBB/BauA 2006a, b
Non-routine cognitive, analytical	High	Increasing	F310, F311, F313, F318
Non-routine cognitive,	High	Increasing	F312, F314
interactive
Routine cognitive	Medium	Falling	F307, F308
Routine manual	Medium/Low	Falling	F304, F305, F306
Non-routine manual	Low	Constant/increasing	F315, F316, F317

^a Alignments are initially based on a factor analysis solution. The variables of the first of the four factors were subsequently aligned to the sub-categories analytical and interactive, respectively. Items marked with "*" were not included in the index because they have high loadings on more than one factor.

^b The question posed was how often the following activities (random order) occur at work -frequently, sometimes or never.  The index is the worker's sum of point values (frequently = 1, sometimes = 0.5, never = 0) divided by the total number of activities in the respective task group.          

F303 Manufacturing, producing of products and goods*
F304 Measuring, testing, quality control
F305 Operating, controlling machines, plants, technical processes
F306 Repairing, maintenance
F307 Buying, providing, selling
F308 Transporting, stocking, posting
F309 Promoting, marketing, public relations*
F310 Organising, planning/preparing work processes
F311 Researching, developing, designing
F312 Training, teaching, tutoring, education
F313 Gathering information, investigating, documenting
F314 Consulting, advising
F315 Entertaining, accommodating, preparing food
F316 Nursing, caring, healing
F317 Securing, protecting, guarding, monitoring, traffic
F318 Working with computers

Table 4: Matching according to task groups

	Under-qualification			Match	Over-qualification			All workers
	Twofold	Skill-based	Formal		Formal	Skill-based	Twofold
Task indicator (0-100, mean values)
Non-routine-cognitive, analytical (F310, F311, F313, F318)	47,5	50,3	56,6	53,6	41,5	43,3	30,8	50,6
Non-routine-cognitive, interactive (F312, F314)	48,4	51,6	56,7	56,5	43,6	50,0	34,0	53,1
Routine-cognitive (F307, F308)	41,1	38,5	39,6	36,5	38,8	39,2	37,0	37,5
Routine-manual (F304, F305,	47,9	46,8	43,6	41,4	36,1	43,4	32,4	41,0
Non-routine manual (F315, F316, F317)	32,6	27,2	23,4	21,3	18,1	21,0	16,5	21,2

Source: BIBB/BAuA Employment Survey 2005/2006, weighted values, own calculations.

Note: the values in the table are mean values of the various matching categories for the task indices and state how much the various matching categories are characterised by these tasks - measured as the frequency with which such activities are exercised. E.g., Workers who are twofold under-qualified perform non-routine analytic activities less frequently than matched workers (47.5 versus 53.6)

Formal and skill-based over-qualification and under-qualification are relevant phenomena of the German labour market, albeit to a limited extent. Over-qualification, as is the case for other countries, plays a greater role than under-qualification. Workers tend to be mismatched in terms of formal credentials rather than with regard to their skills. This means that although relevant proportions of workers are formally under- or over-qualified, their skills or abilities are in fact appropriate for the jobs that they do. We find some striking patterns for relevant subgroups of workers of different socio-demographic and qualification characteristics, among them that workers with an apprenticeship training in the German system of apprenticeship training and university graduates perform equally well in getting a matched job.

This might have a share in explaining why the incidence of mismatching is somewhat lower in Germany than in other countries. Moreover, matching is linked with more non-routine activities and fewer routine activities. Against the background of an increase in formal over-qualification in some European countries, our analyses suggest that mismatching could also rise with changes in the demand for skills in Germany. However, to validate this assumption, the analyses presented here would need to be supplemented by an analysis of longitudinal data.

• AUTOR, D. et al.: The skill content of recent technological change: An empirical exploration. In: Quarterly Journal of Economics, 118 (2003) 4, pp. 1279-1334
• BAUER, T. K.: Educational mismatch and wages. A panel analysis. In: Economics of Education Review 21 (2002) 3, pp. 221-229
• BIERSACK, W. u. a.: Akademiker/-innen auf dem Arbeitsmarkt: Gut positioniert, gefragt und bald sehr knapp. In: IAB-Kurzbericht Nr. 18/2008
• BÜCHEL, F.: The effects of overeducation on productivity in Germany - the firms' viewpoint. In: Economics of Education Review 21 (2002) 3, pp. 263-275
• DALY, M. C.; BÜCHEL, F; DUNCAN GREG. J.: Premiums and penalties for surplus and deficit education. Evidence from the United States and Germany. In: Economics of Education Review 19 (2000) 2, pp. 169-178
• FEHSE, S.; KERST, CH.: Arbeiten unter Wert? Vertikal und horizontal inadäquate Beschäftigung von Hochschulabsolventen der Abschlussjahrgänge 1997 und 2001. In: Beiträge zur Hochschulforschung 29 (2007) 1, pp. 72-98
• GOOS, M.; MANNING, A.; SALOMONS, A.: Job polarization in Europe. In: American Economic Review: Papers & Proceedings 99 (2009) 29, pp. 58-63
• GREEN, F.: Demanding work. The paradox of job quality in the affluent economy. Princeton, N.J.: Princeton University Press 2006
• GREEN, F.; MCINTOSH, S.: "Is there a Genuine Underutilisation of Skills Amongst the Over-qualified?" In: Applied Economics 39 (2007) 4, pp. 427-439
• GROOT, W.; VAN DEN BRINK, H. M.: Overeducation in the labour market. A meta-analysis. In: Economics of Education Review 19 (2000) 2, pp. 149-158
• HARTOG, J.: Over-education and earnings: where are we, where should we go? Economics of Education Review 19 (2000) 2, pp. 131-147
• KONSORTIUM BILDUNGSBERICHTERSTATTUNG: Bildung in Deutschland. Bielefeld 2006
• KORPI, T; TÅHLIN, M.: Educational mismatch, wages and wage growth. Overeducation in Sweden 1974-2000. In: Labour Economics 16 (2009) 2, pp. 183-193
• MCGUINNESS, S.: Overeducation in the labour market. Journal of Economic Surveys, 20 (2006) 3, pp. 387-418
• POLLMANN-SCHULT, M.; MAYER, K. U. 2004: Returns to skills: vocational training in Germany 1935-2000, Max Planck Gesellschaft - eDocument Server. 2004. - URL: http://edoc.mpg.de/ac_oai.pl (Stand: 8.12. 2009)
• ROHRBACH-SCHMIDT, D.; TIEMANN, M.: Qualification and skill-based mismatching in Germany. In: GREEN, F.; KEESE, M. (Eds.): Job tasks, work skills and the labour market (OECD). Paris 2010
• SPITZ-OENER, A.: Technical Change, Job Tasks, and Rising Educational Demands: Looking outside the Wage Structure. In: Journal of Labour Economics 24 (2006) 2, pp. 235-270

 

Daniela Rohrbach-Schmidt
Researcher at the BIBB Research Data Centre (BIBB-FDZ)

Michael Tiemann
Researcher in the "Qualifications, Occupational Integration and Employment" Section at BIBB

Translation from the German original "(Mis-)Matching in Deutschland" (BWP 1/2010)