Compressed, postponed, or disadvantaged? School-to-work-transition patterns and early occupational attainment in West Germany

https://doi.org/10.1016/j.rssm.2016.01.004Get rights and content

Highlights

  • School-to-work transitions in West Germany continue to be rather linear.

  • Yet, the degree of non-linearity of these ideal-typical STWT patterns increased over the cohorts.

  • Higher-educated women experienced higher risks of long-term disadvantage than men.

  • Our findings underscore the merit of looking at the age, and not just durations, in STWT research.

Abstract

We study school-to-work-transition (STWT) patterns and early occupational attainment for five West German birth cohorts. Although these cohorts experienced very different macro conditions, their STWTs were facilitated by Germany's strong vocational education and training (VET) system. The main research question is whether linearity of STWTs differed across and within cohorts. Linearity concerns the normatively expected order of different activity statuses during this life phase. High linearity is ideal-typically defined as entering VET or tertiary education programs after leaving general education, followed by rather direct entry into employment. Non-linear patterns diverge from this ordering or may also include other status activities, like unemployment and inactivity. We use data of the Adult Starting Cohort of the German National Education Panel Study (NEPS) and employ sequence analysis and regression methods. Our analyses reveal that the proportion of young people experiencing the ideal-typical transition patterns increased over the cohorts. Yet, the degree of non-linearity (in terms of number of status activities and status shifts, and some non-employment experience) of these ideal-typical STWT patterns also increased over the cohorts. Moreover, we find strong differences between men and women in early occupational attainment. Higher-educated women in particular had higher risks of long-term disadvantage, whereas men were able to compensate for disadvantages by achieving higher educational attainment and establishing themselves more quickly in the labor market.

Introduction

Linearity is an important characteristic of school-to-work transitions (hereafter: STWTs) in advanced economies (cf. Brzinsky-Fay, 2007, Buchmann and Kriesi, 2011). Linearity concerns the normatively expected order of different activity statuses during this life phase. High linearity is ideal-typically defined as entering vocational education and training (VET) or tertiary education programs after leaving general education, followed by rather direct entry into employment (preferably into skilled jobs) and stable labor market integration. Non-linear patterns diverge from this ordering (e.g., school–employment–VET–employment), or may also include other status activities, like unemployment and inactivity. Thus, the degree of linearity can be assessed by two measures: the number of status activities and the number of shifts between status activities (hereafter: status shifts). Perfectly linear patterns are defined by the three status activities (school, VET/university, employment) and two status shifts (i.e., school–VET and VET–employment). Non-linear patterns are defined by higher number of status activities and status shifts, especially owing to unemployment or inactivity episodes. Additionally, the severity of non-linearity can be assessed by occupational attainment at the (defined) end of STWTs: the higher this attainment of individuals who experienced rather non-linear STWTs, the less severe is this non-linearity.

The discussion on country variation in youth unemployment and NEET rates (youth not in employment, education, and training) often refers to this characteristic of STWTs: unemployment and NEET rates vary considerably between advanced economies. Despite the current crisis, countries with strong VET systems—especially apprenticeship systems (e.g., Austria, Germany, and Switzerland)—show much lower rates than countries with general education systems (e.g., France, Italy, or Spain) (OECD, 2013: 5–6). One of the key arguments why rates are low(er) in these countries is that firm-based training programs facilitate more linear STWT transitions than general education systems (e.g., Brzinsky-Fay, 2007, Buchmann and Kriesi, 2011, De Lange et al., 2014).

This explanation, however, falls short on at least three points. First, most studies do not investigate STWT patterns but rather treat labor market entry as a single event (e.g., Scherer, 2005, Wolbers, 2007). Thus, the linearity of STWT patterns in these countries is often unknown.

Second, most of these comparative studies overlook within-country differences in STWT patterns, often leading to rather ideal-typical considerations of STWT patterns and labor market entry (e.g., Gangl, 2002, Kerckhoff, 2001, Ryan, 2001). Yet STWT patterns may vary across time (cohorts) and within cohorts, for example, by educational attainment or gender.

Third, we can observe a strong association between national youth unemployment rates and the unemployment rates of prime-age workers (aged 25–54) (e.g., Breen, 2005: 131; Gangl, 2002). In 2014, for example, the correlation between these two rates was 0.9 (own calculations based on OECD indicators for 2014).1 This strong correlation questions the simple explanation regarding the impact of VET systems, because low(er) rates could be generated by good economic conditions and not by the education system per se. Breen's (2005) comparative macro-level analysis for the years 1995–1999 suggests that strong VET systems do facilitate lower youth unemployment rates at least to some extent. The time period covered is small, however. The country-comparative study by De Lange et al. (2014) on labor market integration for the period 1992–2008 corroborates Breen's finding. Neither study, however, allows for within-country and within-cohort comparisons over time. To determine whether strong VET systems generate linear STWTs under different economic conditions, we need historical comparisons within countries with strong VET systems. This design enables us to discover whether, and if so for whom, participation in vocational education results in linear STWTs regardless of the economic situation.

The aim of our paper is to overcome these three shortcomings. We therefore study STWT patterns and occupational attainment until age 30 for five West German birth cohorts (born between 1948 and 1977). These cohorts left school between the mid-1960s and the early 1990s and reached age 30 between 1978 and 2007. Thus, they experienced different combinations of macro conditions, but all cohorts’ transitions were facilitated by Germany's strong VET system.

We define the “end” of the STWT period by age rather than duration (i.e., 15 years after leaving school). The reason for doing so is that young people not only stay longer in education (as acknowledged in the duration approach) but also become older. Age receives strong attention in demographic research (e.g., Baizán et al., 2002, Billari and Wilson, 2001) or research on socio-psychological factors (e.g., Elder and Shanahan, 2006, Mortimer, 1994, Shanahan and Mortimer, 1996). Yet it might also be of interest for studying STWTs and early occupational attainment, because being older when making educational decisions and/or entering the labor market could help “speed up” this life passage because of age norms or higher (developmental) maturity (e.g., Mortimer, Oesterle, & Krüger, 2005). Different birth cohorts might therefore achieve similar occupational positions at age 30—even though they had less time between leaving school and reaching 30. Furthermore, delays resulting from educational expansion could be compensated for by higher educational attainment, enabling younger cohorts to achieve higher occupational placements faster. To assess whether this delay not only postpones the entire STWT period but also changes early occupational attainment, and whether this attainment is influenced by maturity and/or age norms, we compare occupational attainment achieved at age 30 and 15 years after leaving general schooling (i.e., when individuals have the same amount of time available for their labor market entry).

In sum, our paper addresses three research questions: First, does the German VET system support linear STWT patterns for men and women—regardless of economic conditions—and does this apply to school leavers with low school attainment as well? Second, does early occupational attainment—as our indicator of the severity of non-linearity of STWT patterns—differ across and within cohorts? Third, does occupational attainment vary when looking at age 30 or 15 years after leaving school?

Section snippets

Macro conditions in West Germany 1960–2000

We start with institutional and historical information on educational expansion, apprenticeship market conditions, and labor market conditions in West Germany between the mid-1960s and the mid-2000s (corresponding to the cohorts of our study, see Table 1).

Theoretical considerations and empirical expectations

The STWT and labor market entry literature provides many well-known theoretical explanations of why and how (relative) educational attainment, labor market structure and conditions, the supply-demand ratio of VET places, and cohort size should affect STWT patterns and early occupational attainment (see also considerations in Section 2). What is less discussed, however, is how these factors interact. As Table 2 displays in a stylized manner, our five cohorts are characterized by very different

Data and methods

For our empirical analyses, we use the data from Wave 3 of Starting Cohort 6: Adults of the German National Educational Panel Study (NEPS) (cf. Allmendinger et al., 2011). This survey is conducted on an annual basis and contains retrospective life-course information on German adults born between 1944 and 1986. Our (unweighted) analysis sample only includes the 6273 West German respondents from five birth cohorts. We are not interested in historical trends but in the empirical investigation of

Findings

We start with some general information on the life period between age 15 and 30. The median school leaving age increased from age 16.3 in the 1950 cohort to age 17.6 in the 1975 cohort—with only minor differences between men and women. In contrast, the median age of first employment increased by only 1.2 years for men but by 2.0 years for women. The main explanation for women's larger delay is their higher increase in university enrollment across the cohorts.

Discussion and conclusions

In the introduction, we argued that STWT research that only looks at labor market entry as a single event is unable to finally prove that vocationally oriented education systems positively influence STWTs. The same is true of cross-sectional country-comparative research. The shortcomings of these approaches include possible endogeneity problems between VET systems and labor market conditions and the tendency to overlook within-cohort differences in STWT patterns. The aim of the paper,

Acknowledgement

For the very constructive comments we thank Marlis Buchmann and the two anonymous reviewers for this paper. The paper was supported by the “College for Interdisciplinary Educational Research. A Joint Initiative of the BMBF, the Jacobs Foundation and the Leibniz Association” (www.ciderweb.org). The paper was first presented at the Symposium “Challenges in the Third Decade of Life in the 21st Century: Individual Development and Health, Social Opportunities and Inequalities” at Castle

References (57)

  • T. Bol et al.

    Educational systems and the trade-off between labor market allocation and equality of educational opportunity

    Comparative Education Review

    (2013)
  • R. Breen

    Explaining cross-national variation in youth unemployment

    European Sociological Review

    (2005)
  • R. Breen et al.

    Bildungsdisparitäten nach sozialer Herkunft und Geschlecht im Wandel – Deutschland im internationalen Vergleich

  • C. Brzinsky-Fay

    Lost in transition? Labour market entry sequences of school leavers in Europe

    European Sociological Review

    (2007)
  • M. Buchmann

    The script of life in modern society

    (1989)
  • M. Buchmann et al.

    Transition to adulthood in Europe

    Annual Review of Sociology

    (2011)
  • G. Carlsson et al.

    Age, cohorts and the generation of generations

    American Sociological Review

    (1970)
  • M. De Lange et al.

    Youth labour market integration across Europe

    European Societies

    (2014)
  • R.A. Easterlin

    Birth cohort and fortune

    (1980)
  • G.H. Elder

    Children of the great depression

    (1999)
  • G.H. Elder et al.

    The life course and human development

  • B.S. Everitt et al.

    Cluster analysis

    (2011)
  • M. Gangl

    European patterns of labour market entry

    European Societies

    (2001)
  • M. Gangl

    Changing labour markets and early career outcomes

    Work, Employment & Society

    (2002)
  • C. Hakim

    Key issues in women's work

    (1996)
  • A. Hall et al.

    Erfolgreich im Beruf? Duale und schulische Ausbildungen im Vergleich. BIBB Report 2/2014

    (2014)
  • S. Hillmert

    Labour market integration and institutions: An Anglo-German comparison

    Work, Employment & Society

    (2002)
  • Cited by (65)

    • The role of family social capital in school-to-work transitions of young adults in Germany

      2022, Journal of Vocational Behavior
      Citation Excerpt :

      Since we considered the whole 60-month period after leaving general education, some gaps exist in the young people's occupational biographies due to missing status information for those particular months. In order to reduce the complexity of the individual transition patterns, we generated different types of transition patterns using a combination of sequence and cluster analyses – a common analytical strategy for examining STWT in a life course perspective (Brzinsky-Fay, 2007, 2014; Brzinsky-Fay & Solga, 2016). We used the optimal matching algorithm proposed by Brzinsky-Fay et al. (2006) which compares the individual STWT sequences with each other, generating a similarity measure which is used by the Ward algorithm to create clusters by minimising the within-group difference while maximising the between-group difference.

    • The impact of the abolition of compulsory service on life satisfaction

      2022, Research in Social Stratification and Mobility
      Citation Excerpt :

      Additionally, compulsory services cause deviations from the standard transition from school to (vocational) training3 and into stable employment (Brzinsky-Fay & Solga, 2016; Scherer, 2005, 432), i.e. career disruptions. Such disruptions may result in short- and long-term disadvantages, especially when they occur during a stage in the life course that is accompanied by major changes, such as the transition from school to work, moving out of the parental home, or starting a family (Brzinsky-Fay & Solga, 2016; Manzoni & Mooi-Reci, 2011). Thus, we expect compulsory services to lower life satisfaction of young men before and after the draft because of uncertainty regarding the future and career disruptions (Elder, 1987; London & Wilmoth, 2006).

    • Pathways, labor markets and tertiary education systems

      2022, International Encyclopedia of Education: Fourth Edition
    View all citing articles on Scopus
    View full text