The BIBB-IAB qualification and occupational field projections (QuBe project) (HELMRICH and ZIKA 2010, HELMRICH et al. 2012, MAIER et al. 2014a, MAIER et al. 2014b, ZIKA et al. 2012) are a coordinated projection of demand and supply on the basis of jointly defined occupational fields and data generations. It is the comparison of the labour demand and supply at the occupational level in particular which makes the BIBB-IAB qualification and occupational field projections different from previous studies (e.g. BLK 2002, BONIN et al. 2007, DOSTAL 2002) which were limited mainly to the levels of qualifications, individual sectors or abstract areas of activity. Projections at occupational level were not feasible due to a lack of classifications and characteristics which enable the demand and supply to be attributed in a manageable way.
On the basis of data from the Microcensus surveys of the Federal Statistical Office, figures have been compiled since 1996 for persons in employment by occupation practised and industrial sector association and for the economically active population by skills level, age and gender. Since the survey year 2005, data on the learned occupation of the economically active population has also been available. This allows the flexibility of the economically active population to be taken into account in the calculation. The supply of the economically active people with a learned occupation can therefore be translated into a potential labour supply by occupation practised.
At the qualification level, a distinction is made between four levels following the ISCED classification framework, namely: People with no formal vocational qualification (ISCED 1,2,3a); People with vocational training (ISCED 3b, 4); People with a master craftsman, technical engineer, vocational school qualification (ISCED 5b); People with an academic degree (ISCED) 5a, 6). People currently in education and training (e.g. school, training, degree) are shown as a separate category.
At the occupational level, the BIBB occupational field classifications were applied to (TIEMANN et al. 2008) both the demand and the supply side. These consist of 54 occupational fields which are grouped at the level of the occupational categories (3-digit codes from the classification of occupations 1992 (KldB 92)) and in each case have comparable job characteristics and branch dominance.
The focus on a vocational flexibility matrix can be regarded as a unique aspect. This reflects the extent to which people with a specific learned occupation stay in this occupation during their working life and how often they work in other occupations.
In the first projection wave (HELMRICH und ZIKA 2010) it was already possible to assess labour supply against labour demand at the individual level taking into account this flexibility, but this assessment was expanded in the second project wave at the volume of work level (ZIKA et al. 2012). For this purpose, the potential volume of work was also calculated from the Microcensus by gender, age, qualification and occupational field. This describes the extent to which the economically active population are prepared to work more than they currently do. The potential volume of work is also a unique feature of the QuBe project.
In the third wave of the QuBe project (MAIER et al. 2014a, MAIER et al. 2014b), the interactions between the demand and supply side are modelled by taking into account salaries specific to the occupational field. In this process, the movement in salary is dependent on potential workforce surpluses and shortages in the occupational fields, and the economically active population respond to these salary changes according to their vocational flexibility. However, this dynamic between the supply and demand side can only be shown where it has been possible to determine responses in the past. This empirically supported approach is the basis for each and every baseline projection in the QuBe project (siehe MAIER et al. 2014b).
The international comparability of project results largely depends on two main factors: The data source selected and the chosen classifications. The approach of the QuBe project relies on a detailed evaluation of the Microcensus. The Microcensus also represents the German form of the Labour Force Survey. International studies which draw on the labour force survey (e.g. the “forecasting skill demand and supply” project by CEDEFOP) therefore arrive at similar trends to the QuBe project (MAIER 2011). In principle, a comparison is possible by qualification level with the aid of the ISCED classification. The industrial sector classification WZ08 used in the third wave is also compatible with the NACE Rev 2 used internationally, which means that comparability of persons in employment by industrial sector becomes possible. Comparability by occupation is somewhat limited, but still possible. For example, the conversion of the national occupational classifications for 1988 and 1992 to the new 2010 occupational classification also results in a break in the time series for international comparisons (ISCO-2008). All projects which use the new occupational classifications must therefore find a way to be recoded between the time series used in the earlier classifications and the new surveys in the new classifications. This type of recoding is also possible with the BIBB occupational fields.
No record is made of public officials, freelancers and unpaid family workers in the statistics for employees subject to social security deductions from the Federal Employment Agency. In addition to this, for employees subject to social security deductions who have completed training in the dual system, information is only available regarding the learned occupations at the level of occupational classification (3-digit codes from the classification of occupations 1988).
The education and training data from the official statistics (university, vocational education and training) certainly report the training participants and graduates from education and training (and therefore the new entrants from the education and training system into employment), but say nothing about labour force participation and gainful employment or about people without formal vocational qualifications.
The only official representative statistic about the population and the job market in Germany is the Microcensus. The information it provides includes the population structure, the economic and social situation of the population as well as occupations. Job hunting is also included as well as representative information regarding the subject field of the highest level vocational qualification completed for all economically active persons in Germany across all education and training sectors.
However, due to a lack of alternatives, the salary data for employees subject to social security deductions has been used as salary information in the QuBe project since the third wave (MAIER et al. 2014b).
The Microcensus underestimates the number of persons in dependent employment, in particular those in marginal employment.
The under recording of people in dependent employment in the Microcensus (for example, approx. 2 million people in 2005) is problematic compared to the calculation of persons in dependent employment by the Federal Statistical Office (KÖHNE-FINSTER und LINGNAU 2008). The material difference was sufficient reasons to adjust the data by means of an iterative marginal total adjustment procedure. The figures for people in dependent employment from the Microcensus were adjusted for the figures for persons in dependent employment from the VGR. Here the marginal totals of the total population by age, gender, level of education (ISCED) and learned occupation in the Microcensus were kept constant. This restrictive approach guarantees that the central structure of the Microcensus is retained for the projections.
No explicit assumptions were made with regard to economic growth and the trends in productivity. Instead due to the large number of relevant parameters for recording the complexity of the economy, a model based projection system was used for forecasting – the so-called IAB/INFORGE model.
The IAB/INFORGE model distinguishes between 63 economic sectors based on the 2008 industrial sector classification.
The IAB/INFORGE model is an econometric forecast and simulation model for the Federal Republic of Germany which is highly disaggregated by areas of production and commodity groups and which has been developed by the Institute of Economic Structures Research (GWS mbH) (MEYER et al. 2007, SCHNUR und ZIKA 2009). It is particularly effective due to its integration in an international model network. The model is constructed based on the "bottom up" and "full integration" principle. "Bottom up" implies that the individual sectors of the economy are modelled in great detail and the macroeconomic variables are created by aggregation in the model structure. "Full integration" stands for complex and simultaneous modelling of the inter-industry supply integration and the creation and distribution of income, the redistribution activity of the state and the application of income by private households.
Labour supply is projected (the economically active population (persons in dependent employment and the unemployed)) with learned occupations both for new supply as well as for the existing economically active population, and this is carried out at four qualification levels under the ISCED classification and under a separate category for persons in education and training as well as for 54 occupational fields.
The labour supply projections are based on several transitional models by which, starting with an initial population, future economically active populations are modelled using inflows and outflows. The new supply is adjusted in the projection period as far as possible to the benchmark figures from the Standing Conference of Ministers of Education and Cultural Affairs projection for school children and graduate numbers as well as degree entrants (KMK 2012, KMK 2013). The reduction in school time at "Gymnasien" (German grammar schools) and the resulting double cohorts of graduates with qualifications for higher education are therefore taken into account. This projection of the number of degree entrants is a status quo projection with the rate of degree entrants held constant. The number of degree entrants therefore increases due to the rising number of school leavers in the secondary sector II with higher education entrance qualification and is reflected accordingly in the rising new supply of economically active persons with an academic degree. The future qualification structure of the economically active population however is also increasingly determined by individuals leaving working life. These are predominantly qualified in the intermediate qualification area. The population projected by qualification level using new supply and the remaining population participates to varying degrees in working life depending on age, gender and qualification level.
Version 1W1 of the 12th coordinated population projection was used for the first two projection waves with a slightly rising life expectancy, a positive net immigration of 100,000 persons annually and a constant birth rate of approx. 1.4 births per female. According to preliminary results of the migration statistics from the Federal Statistical Office (STATISTISCHES BUNDESAMT 2013), the net migration for 2012 rose to approximately 369,000. For the third projection wave we are therefore using version 1-W2 of the 12th coordinated population projection which assumes a further increase in net migration to 200,000 individuals by 2020.
While even more emigration than immigration was recorded in Germany in 2008 and 2009, constantly rising migration surpluses have been recorded since 2010. In 2012 the number was 370,000 people. According the version 1-W2 of the population projection currently being used, the net immigration for 2012 is only 80,000 people. It must be assumed that currently the relatively high net immigration to Germany is a response to the crisis in some European countries (BERTOLI et al. 2013) and that, in view of the imminent demographic challenges facing Germany, the migration movement will stabilise over the long term at a higher level than the average over the last decade.
With regard to the qualification structure of the net immigration, it is assumed that it reflects the net new supply from the education and training system (neutral hypothesis). It is certainly recognized, for example, that the proportion of university graduates amongst the immigrants coming to Germany has significantly increased (SEIBERT und WAPLER 2012), however there is no statistical system which records the qualifications of these migrants. Studies examining at least sections of the migration, largely support the neutral hypothesis (ETTE und SAUER 2010).
These are predominantly qualified in the intermediate qualification area. The population projected by qualification level using new supply and the remaining population now participates to varying degrees in working life depending on age, gender and qualification level. Trend estimates are used to update the labour force participation rates which represent economically active persons in the population of the same age, gender and qualification level. Essentially, the labour force participation rate mainly indicates increases among women and the elderly in past years. These also continue in this way in a weaker form for the projections. The adjustment of the statutory retirement age from 65 to 67 years is taken into account using a gradual increase in the labour force participation rate in the over 60s age group.
The potential economically active population is the sum of the economically active population and the so-called hidden reserves. The hidden reserves are the part of the population which does not voluntarily participate in working life but, who, under certain circumstances, would consider participation in the labour force. However, it must be assumed that due to the rising labour force participation rates the hidden reserves will slowly decline.
The potential volume of work is a hypothetical construct which states the actual size of the labour supply measured in hours. In order to calculate this construct and using a 1 percent sample of Germany's resident population, the Microcensus uses the number of desired weekly working hours, provided these are above the working hours actually worked on a regular basis (ZIKA et al. 2012: 8).
The qualification levels were selected according to the ISCED classification to make comparability with international projection models possible. To start with, only four groups of qualifications were distinguished which were most significant in terms of numbers and with regard to education background. Persons currently in education and training (e.g. school pupils, trainees, students) are shown as a separate category.
In the first stage, these four qualification levels were used to differentiate because they represented the relevant qualification levels for the job market.
Individuals with the higher education entrance qualification have a far better chance on the job market than individuals without a vocational qualification who do not have the higher education entrance qualification (BRAUN et al. 2012).
Individuals in education and training are counted as persons in dependent employment, although they are not assigned a learned occupation, but instead the category “in education and training”. This also applies if they have already obtained a vocational qualification in a trade.
The BIBB occupational fields were created on the basis of the 2006 Employee Survey conducted by the Federal Institute for Vocational Education and Training (BIBB) and the Federal Institute for Occupational Safety and Health (BAuA) as well as the 2004 Microcencus based on the 3-digit codes from the classification of occupations 1992. Here, occupations were defined by means of clustering occupations with similar activities. As a second priority, the respective sector focus of persons in dependent employment was also used to create the groups. Persons in dependent employment in an occupational field therefore exhibit a high level of homogeneity in terms of activities within the same occupational field and a high level of heterogeneity compared to other occupational fields.
In the BIBB-IAB qualifications and occupational field projections, the occupation is regarded as the key element between training and the employment side. It links together the abilities and skills which are obtained in education and training or in a degree with the activities performed in working life. But because the occupational fields were created using the activity characteristics, they can also be interpreted as activity clusters.
This has not so far been explicitly addressed in the projections, but is part of other investigations. However, the fact that qualification requirements change within occupations has been taken into account. In the future it will be possible to incorporate occupational changes such as these into the analyses based on longer time series which will then be available.
The learned occupation describes the abilities and skills which are learned in the education and training system. These skills deliver the capacity to perform certain employment activities. By recording both the learned and practised occupations, it is possible to show which learned skills (or occupations) lead to which employment activities (practised occupations). This therefore makes it possible to represent the opportunities and risks resulting from learning a specific occupation.
In the Microcensus, which is used to calculate the learned occupation, only the highest level vocational qualification is used. Previous diplomas and qualifications are therefore not incorporated in the calculations. However, longitudinal studies show that an individual's additional qualifications are often undertaken in a similar occupation to the first (JACOB 2004).
The frequency with which a different occupation is practised for a specific learned occupation is expressed in the vocational flexibility matrix. However, the reasons for the change from the learned occupation to another occupation are not reflected in this vocational flexibility matrix. There can be varied reasons for this. A change might be made, for example, due to improved career options or to the threat of job losses, or job losses which have already occurred in the learned occupation (z.B. HALL 2010). Because the occupational fields used in the projections already combine occupations involving similar activities, statistical artefacts from the change in occupation are generally avoided at a small scale level by means of the occupation descriptions used (e.g. a change from painter and decorator to painter and decorator for conversions).
It must be noted that the learned and practised occupations in the Microcensus are based on information volunteered by the sample. The name given to a person's own practised occupations may therefore by influenced both by the job description as well as by the features of the activities of the occupation practised. If a mathematician works as a bank clerk, then it is ultimately the qualifications of a bank clerk which are needed and not necessarily those of a mathematician. It is conceivable that MINT qualifications are often in strong demand in other areas. However, if in this instance other occupational titles with corresponding training occupations or specialisms are relevant, the appointment of an individual with different learned occupation to such a practised occupations is not a necessity but an expression of flexibility.
Some learned occupations have greater flexibility, others are more rigid.
This is not a new finding in itself, but this is part of the projections for all occupational fields. Flexibility is lower for occupations which are professionally regulated and with a restrictive access requirement than for occupations involving more straightforward activities and which do not have rules of professional conduct.
The flexibility matrix describes the processes of moving between a learned occupation and a practised occupation and describes these only. The reasons for a change are not specified.
For the time being, changes in occupation are neutral in this analysis. In reality, there are changes in occupation which are advantageous for persons in dependent employment and those which are out of necessity and tend to be associated with negative consequences for the person concerned. However, research into the causes of this is not a primary aim of the projections.
We update historical trends observed in the education and training system and on the job market in our calculations. If interdependencies between different variables can be observed historically, then we take these interdependencies into account for the future. If no trends can be identified for certain variables, then we retain the status quo in the updates. The results are intended to reveal the path we are currently on and where undesirable developments might have to be corrected. We do not assume that the results will actually occur as they are. In each case, the individual projections themselves do indeed highlight realistic development paths, however they are based solely on existing findings from the past. The adjustment responses between labour demand and supply at occupational level (salary increases and changes in vocational flexibility) which were considered in the third wave are also only modelled in those instances where that can be empirically proven. However, future behaviour can also differ from past behaviour or situations might arise which could not have been perceived in the past (e.g. specialist skills shortages in certain occupations). The results therefore also have the nature of an if-then statement and do not represent a forecast in the sense of a fixed and unchangeable depiction of the future.
Projections are neither factual targets nor are they inevitable, instead they are a magnifying glass for what appears to be most likely at the moment. And if a scientifically proven and conscientiously prepared projection does not actually occur, then this does not make it bad because it may provide guidance and / or have provoked changes in behaviour. And if it has provoked such changes in behaviour, then it has not only fulfilled its purpose, but, of course, it has also changed its own projection bases. It is therefore, generally speaking, virtually impossible to validate projection results. For all projections it is necessary to extrapolate this historical development into the present and on into the future. We consider, what is likely to be the case if the future develops in the way things have developed in the past, incorporating what we already believe we know about the future. Projections are nothing more than a description of that which would occur if the current trends remain unchanged.
This occurs subject to a range of assumptions, e.g. regarding economic growth, population development, labour force participation rates, etc. If a parameter changes, then it is inevitable that the result will also change.
The potential creation of self-fulfilling prophecies is regarded as a particular risk with labour market projections, for example, in the form of so called pork-cycles. The engineering and teaching occupations are examples of such cyclical job markets in Germany. The cause of trends like these is the delayed actions of the providers of specialist skills.
It is only possible to prevent such exaggerated responses to known projection results if the projections are repeated at regular intervals. It is only in this way that the impacts of change to the initial situations can be taken into account and if necessary counteracted.
No, a development such as this is definitely not possible. The number of persons in dependent employment can never be lower than the economically active population (persons in dependent employment + unemployed). This will always result in adjustment processes of some form or another (e.g. reduction in demand, increase in supply) which compensates for this imbalance.
BERTOLI, Simone; BRÜCKER, Herbert; FERNÁNDEZ-HUERTAS MORAGA, Jesús: The European crisis and migration to Germany expectations and the diversion of migration flows. IZA Discussion Paper No. 7170. Bonn 2013.
BLK, Bund-Länder-Kommission für Bildungsplanung und Forschungsförderung: Zukunft von Bildung und Arbeit. Perspektiven von Arbeitskräftebedarf und -angebot bis 2015. Bonn 2002
BONIN, Holger u.a.: Zukunft von Bildung und Arbeit. Perspektiven von Arbeitskräftebedarf und -angebot bis 2020. 2007
BRAUN, Uta u.a.: Erwerbstätigkeit ohne Berufsabschluss - Welche Wege stehen offen? In: BIBB-Report, 17/12 (2012), p. 1-10
DOSTAL, Werner: Die IAB-Prognos Tätigkeits- und Qualifikationsprojektionen. In: BUNDESINSTITUT FÜR BERUFSBILDUNG (Hrsg.): Veränderte Arbeitswelt - veränderte Qualifikationen. Wechselwirkungen zwischen Arbeitsmarkt und Bildungsstrukturen. 2002, p. 53-66
DROSDOWSKI, Thomas; WOLTER, Marc Ingo: Entwicklung der Erwerbspersonen nach Berufen und Qualifikationen. In: HELMRICH, ROBERT; ZIKA, GERD (Hrsg.): Beruf und Qualifikation in der Zukunft. BIBB-IAB-Modellrechnungen zu den Entwicklungen in Berufsfeldern und Qualifikationen bis 2025. Bielefeld 2010, p. 125-152
ETTE, Andreas; SAUER, Lenore: Auswanderung aus Deutschland: Daten und Analysen zur internationalen Migration deutscher Staatsbürger. Wiesbaden 2010
HALL, Anja: Wechsel des erlernten Berufs. Theoretische Relevanz, Messprobleme und Einkommenseffekte. In: Zeitschrift für Berufs- und Wirtschaftspädagogik, Sonderheft 24: Berufsforschung für eine moderne Berufsbildung - Stand und Perspektiven (2010), p. 157-174
HELMRICH, Robert; ZIKA, Gerd (Hrsg.): Beruf und Qualifikation in der Zukunft. BIBB-IAB-Modellrechnungen zu den Entwicklungen in Berufsfeldern und Qualifikationen bis 2025. Bielefeld 2010
HELMRICH, Robert u.a.: Engpässe auf dem Arbeitsmarkt: Geändertes Bildungs- und Erwerbsverhalten mildert Fachkräftemangel. Bonn 2012.
JACOB, Marita: Mehrfachausbildungen in Deutschland. 2004
KALINOWSKI, Michael; QUINKE, Hermann: Projektion des Arbeitskräfteangebots bis 2025 nach Qualifikationsstufen und Berufsfeldern. In: HELMRICH, ROBERT; ZIKA, GERD (Hrsg.): Beruf und Qualifikation in der Zukunft. BIBB-IAB-Modellrechnungen zu den Entwicklungen in Berufsfeldern und Qualifikationen bis 2025. Bielefeld 2010, p. 103-124
KMK, Statistische Veröffentlichung der Kultusministerkonferenz: Vorausberechnung der Studienanfängerzahlen 2012-2025 - Fortschreibung. Dokumentation Nr. 197. Berlin 2012.
KMK, Statistische Veröffentlichung der Kultusministerkonferenz: Vorausberechnung der Schüler und Absolventenzahlen 2012-2025. Dokumentation Nr. 200. Berlin 2013.
KÖHNE-FINSTER, Sabine; LINGNAU, Andreas: Untersuchung der Datenqualität erwerbsstatistischer Angaben im Mikrozensus - Ergebnisse des Projekts "Nachbefragung im Mikrozensus/LFS". In: Wirtschaft und Statistik, 12 (2008), p. 1067-1088
MAIER, Tobias: Methods and Results of Skills Demand and Supply Forecasting - the Case of Germany. Building on the skills forecasts. Comparing methods and applications. Thessaloniki 2011.
MAIER, Tobias u.a.: Engpässe im mittleren Qualifikationsbereich trotz erhöhter Zuwanderung. In: BIBB-Report 23/14 (2014a)
MAIER, Tobias u.a.: Löhne und berufliche Flexibilitäten als Determinanten des interaktiven QuBe-Arbeitsmarktmodells. Ein Methodenbericht zur Basisprojektion der 3. Welle der BIBB-IAB Qualifikations- und Berufsfeldprojektionen. In: Wissenschaftliche Diskussionpapiere Heft 148. Bonn 2014b.
MEYER, Bernd u.a.: National economic policy simulations with global interdependencies. A sensitivity analysis for Germany. In: Economic Systems Research, 19 (2007) 1, p. 37-55
SCHNUR, Peter; ZIKA, Gerd: Das IAB/INFORGE-Modell. Ein sektorales makroökonomisches Projektions- und Simulationsmodell zur Vorausschätzung des längerfristigen Arbeitskräftebedarfs. Nürnberg 2009
SEIBERT, Holger; WAPLER, Rüdiger: Zuwanderung nach Deutschland: Aus dem Ausland kommen immer mehr Akademiker. In: IAB-Kurzbericht, 21/2012 (2012)
STATISTISCHES BUNDESAMT: Bevölkerung und Erwerbstätigkeit - Vorläufige Wanderungsergebnisse 2012. Fachserie 1 Reihe 2. Wiesbaden 2013.
TIEMANN, Michael u.a.: Berufsfeld-Definitionen des BIBB auf Basis der KldB1992. In: Wissenschaftliche Diskussionspapiere Heft 105. Bonn 2008.
ZIKA, Gerd u.a.: In der Arbeitszeit steckt noch eine Menge Potenzial. Qualifikations- und Berufsfeldprojektionen bis 2030. In: IAB-Kurzbericht, 18/2012 (2012), p. 1-12