Occupations and survey research: methodological and substantive applications on the occupation-inequality link 1 |
|
| Chair | Dr Daniela Rohrbach-Schmidt (German Federal Institute for Vocational Education and Training ) |
| Coordinator 1 | Professor Christian Ebner (University of Cologne) |
Recently, there has been an increase in studies that aim to identify the role of occupations and occupational characteristics for individual labor market placement. For Germany with its strong occupation-specific system of vocational education and training (VET), it can be assumed that graduates’ chances of finding vertically and horizontally matching positions are in particular related to the occupation of training.
In this paper, I focus on early employment trajectories of graduates of apprenticeship training in Germany. I explore how training occupations differ in their structural characteristics and analyze how these characteristics translate into unequal labor market chances for graduates. I focus on four characteristics of training occupations: training costs for apprenticeship firms, degree of occupational closure, specificity of acquired skills, and gender segregation.
I put a special focus on the question how sustainable effects of training occupation characteristics are, i.e. for how long graduates’ labor market chances are influenced by these characteristics: Do they have an impact only directly after graduation on the first position in the labor market - and further employment patterns are then primarily determined by this first position? Or do structural characteristics of training occupations also have an independent effect on further employment patterns – leading to unequal chances to remain in favorable or to leave unfavorable first positions?
To study this question, I use retrospective life course information from the adult cohort of the German National Education Panel Study (NEPS) for a sample of individuals from West Germany who graduated from apprenticeship training between 1974 and 2004 in the most common training occupations. Observing these graduates during the first five years after apprenticeship, I employ techniques of sequence analysis to identify distinct employment patterns. In order to access the influence of occupational characteristics on the observed patterns, I construct indicators at the occupational level from data of the Employment Surveys of the Federal Institute for Vocational Education and Training (BIBB) and the Institute for Employment Research (IAB) from 1979 to 1999 and from data of the BIBB’s studies on Costs and Benefits of Apprenticeship Training for Firms from 1980 and 2000. These measures are linked to the training occupations in the individual-level data using the German Classification of Occupations 1988 (KldB88).
First results indicate that graduates’ employment patterns in the first five years in the labour market are quite stable and that the first position obtained after apprenticeship training has a strong impact on these patterns. Structural characteristics of training occupations are primarily decisive for the chances of finding horizontally and vertically matching first positions; however, especially average training costs and gender composition of an occupation also have an independent effect on further employment patterns.
A decisive transition in life is the passage from training to work. In Germany there has been a significant increase in precarious working arrangements across all age cohorts. To what extent does this affect young skilled employees who have completed vocational education and training? What is the influence of the training occupation? Are there differences in these factors for young men and young women or other individual factors?
The occupational integration of individuals is often analysed and classified as being integrated, precarious or disaffiliated (e.g. the zones of integration by Robert Castel). For cross-sectional data referring to the occupational status there are some crucial, more or less generally accepted aspects for the assignment of a particular occupational status to one of the three types of occupational integration.
To evaluate histories in longitudinal studies we have developed two indices, one for integration and one for disaffiliation. Both indices include parameters for the total time of statuses, assigned to a particular zone of occupational integration in the observation period, the duration of continuous statuses and the number of employment interruptions. If there is overall sufficient information regarding occupational integration (unlike histories characterized by family work, further training etc.) for the integration index as well as for the disaffiliation index the parameters for integration respectively disaffiliation should be dominant for defining a history as integrated or decoupled. Otherwise, if none of the parameters is dominant, histories are assigned as precarious. By this it is possible to classify job histories with regard to occupational integration and compare them between cohorts or even different datasets.
With multilevel analysis it is possible to analyse the probability of being classified as integrated, precarious or disaffiliated with the training occupation on a structural level. We also analyse context effects and additional individual effects like school leaving certificates or gender.
As an example we evaluate statistically a three-year period of young people entering the Labour market after completing vocational training, using data of the German Sample of Integrated Labour Market Biographies (SIAB) 2008.
International comparative research has identified several dimensions of national educational systems, which matter for young people’s educational and labor market outcomes, including skill mismatch (Breen, 2005; Levels, van der Velden, & Di Stasio, 2014; Van de Werfhorst & Mijs, 2010; Wolbers, 2007). However, research concerning institutional variation within national education systems and its influence on skill mismatch is scares. Our aim is therefore to investigate whether differences in institutional characteristics between the approximately 230 Swiss upper-secondary VET programs such as, for example, the ratio of occupation-specific versus general knowledge and exam standardization - affect skill-inadequate employment of young diploma holders. This question is particularly relevant for Switzerland, where two thirds of all young people enter upper-secondary VET.
Our key assumption is that differences in the institutional characteristics of VET programs lead to considerable variation in their capacity for skill and competence development (W. Müller & Shavit, 1998). We argue that the degree of exam standardization as well as the vertical and horizontal differentiation of the training occupations leads to higher skill- and competence-homogeneity of the student group. Student homogeneity increases the signaling power of the diploma. Further, a large share of occupation-specific training is likely to decrease the necessity of later on-the-job training and increase initial productivity of VET diploma holders after labor market entry. According to this theoretical framework we hypothesize that a high share of occupation-specific training, high levels of horizontal and vertical differentiation and high exam standardization decreases the risk of skill-mismatch at labor market entry.
The analyses are based on a sample of 3000 VET diploma holders who have graduated within the last five years and who have no further education and training (Swiss labor force survey). Employment is considered to be skill-inadequate if the trained and the current occupation have different two digit codes of the Swiss Standard Classification of Occupations 2000. The individual level data is used in combination with collected data on institutional characteristics from training ordinances and curricula. Measuring vocational specificity, we include the share of occupation-specific lessons and training in school, inter-company courses and in the training firm. An index for horizontal differentiation measures the sorting of students into different subjects, whilst vertical differentiation distinguishes between tracks leading to different diplomas. Regarding standardization, we consider if exams are regionally centralized and what proportion of the final grades are based on teacher’s evaluation during the training year as opposed to final exams. Control variables are age at graduation, time since graduation, gender, migration background, number of diplomas, currently in further education, firm size and labor market opportunities. To account for the clustering of individuals into education programs we use logistic multilevel regression models. Descriptive results show that programs vary considerably in their institutional characteristics. Furthermore, the findings support the hypotheses that institutional characteristics of vocational education and training programs matter for skill-mismatch.
In this presentation, I will analyze the labor market impact of non-native workers on native workers’ unemployment risks in Germany. The novelty of this approach is that it draws from closure theory and addresses the occupational structure of the German labor market. There are two advantages to this strategy: First, I am able to include the impact of closure processes that determine the job security experienced by the native incumbents of these occupations. Second, changing the focus of analysis to the occupational level also addresses the fact that the German labor market consists of employees whose qualifications and skills are related to occupation-specific task fields. Hence, employers’ access to suitable (non-native) candidates is restricted to those who have the required occupational credentials and skills. The occupational fragmentation of the German labor market should thereby work in favor of native employees. High shares of non-native occupational incumbents do not generally worsen the employment security of the native workforce. However, some groups within the native workforce might face an increased unemployment risk (substitution), whereas others might benefit and experience a lower unemployment risk (complementation).
Using the full 1 percent population sample of the German Microcensus, I apply a two-step multilevel model that accounts for the occupational segmentation of the German labor market. The two-step approach allows me to analyze the dichotomous outcome variable employment/ unemployment without introducing substantial biases into the analyses, due to occupations without unemployed incumbents. Conventional multilevel analyses would drop these individuals/occupations because they do not vary on the individual level. Exact logistic regressions allow me to circumvent the low-event bias that occurs in occupations where very few individuals are unemployed.
Empirically, the presentation lends credit to the assumption that the occupational segmentation of the German labor market affects the unemployment risk of the native workforce. The share of non-native workers has no effect on the native occupational members’ unemployment risks. It also shows that processes of occupational closure create additional social and legal boundaries around occupational labor markets and therefore determine the job security experienced by the native incumbents of these occupations. A low supply of occupation-specific credentials reduces the unemployment risk, whereas the incumbents’ unemployment risk increases if occupations award occupational credentials in an inflationary manner. I am also able to provide evidence that task niches decrease the incumbents’ unemployment risk, whereas native incumbents of occupations that are not characterized as a task niche (with common sets of tasks) are more likely to be substituted by non-native workers and to become unemployed.
In sum, the empirical results indicate that the occupational fragmentation of the German labor market works to the advantage of its native employees. High shares of non-native occupational incumbents do not generally worsen the employment security of the native workforce. However, some groups within the native workforce face an increased unemployment risk, whereas others benefit and experience a lower unemployment risk depending on the commonness and uniqueness of their occupation-specific tasks.
Wage returns to education are important social parameters, especially in vocationally oriented labour markets such as Germany. The literature provides competing hypotheses on how educational returns might have developed over time given that decisive changes in the supply of skills (educational expansion, demographic shifts) and the demand of skills (technological change) have taken place. One assumption that gains increasing attention in that context is that a positional model of education becomes more important (Bol, 2015). In this model, individual labor market rewards would not primarily depend on absolute skill levels, but rather on worker’s relative position in the labor market.
To date, few studies have comparatively analyzed the explanatory power of different theories on changes in the value of education. We apply the question of changing wage returns to education to the case of Germany and thus aim to makes several contributions to the literature. First, we present estimates for the wage returns to educational qualifications for West-German male workers over the period of 1976 to 2010 using the high-quality Sample of Integrated Labour Market Biographies (SIAB), a two percent random sample of the Integrated Employment Biographies. Second, following Ortiz and Rodriguez-Menés (2015), from human capital theory, credentialism, skill-biased technological change and the routine-biased technological change, the paper derives competing hypotheses on the development of educational returns over time and tests these hypotheses by exploiting a random coefficient framework. More specifically, we take advantage of the nested structure of the data, i.e. workers nested in occupations to show that wages as well as wage premiums of education significantly vary between 3-digit occupations. We then advance the Mincer-model by substantially explaining variables at the occupation level, i.e. among other, measures for the skill composition of occupations, their degree of closure and educational specificity, to study their direct and indirect (via educational returns) wage effects.
First results seem to support previous findings on rising returns to tertiary education and constant to falling returns to vocational education in Germany, and thereby contributing to rising wage inequality. Moreover, the models show that wage premiums for different educational levels strongly depend on occupations over the whole period, resulting in equally high or even higher payoffs to upper secondary vocational education than tertiary education in some occupations even in the most current years of observations. Results for mediating effects of occupations still stand out, but should be available at the time of the conference.
References:
Bol, T. (2015). Has education become more positional? Educational expansion andlabor market outcomes, 1985–2007. Acta Sociologica, 58(2), 105–120.
Ortiz, L., & Rodriguez-Menes, J. (2015). The Positional value of education and its effect on general and technical fields of education: educational expansion and occupational returns to education in Spain. European Sociological Review (24), 1–22.