Smooth Path or Long and Winding Road? How Institutions Shape the Transition from Higher Education to Work

Leuze, Kathrin Book — Published Version Smooth Path or Long and Winding Road? How Institutions Shape the Transition from Higher Education to Work Provided in Cooperation with: WZB Berlin Social Science Center Suggested Citation: Leuze, Kathrin (2010) : Smooth Path or Long and Winding Road? How Institutions Shape the Transition from Higher Education to Work, ISBN 978-3-86388-284-6, Budrich UniPress, Leverkusen-Opladen, https://doi.org/10.3224/94075542 This Version is available at: https://hdl.handle.net/10419/251573 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by-nc-nd/3.0/ Kathrin Leuze Smooth Path or Long and Winding Road? Kathrin Leuze Smooth Path or Long and Winding Road? How Institutions Shape the Transition from Higher Education to Work Budrich UniPress Ltd. Opladen & Farmington Hills, MI 2010 You must not circulate this book in any other binding or cover and you must impose this same condition on any acquirer. A CIP catalogue record for this book is available from Die Deutsche Bibliothek (The German Library) © This work is produced under Creative Commons lincense: http://creativecommons.org/licenses/by-nc-nd/3.0/de/ The work may be distributed freely in a non-commerical setting. For commercial use, please contact Budrich UniPress This book maybe downloaded from www.budrich-unipress.com A hard copy maybe ordered directly from Budrich UniPress or your local bookstore. Page numbers in print and online version are identical ISBN 978-3-940755-42-1 / DOI 10.3224/94075542 Budrich UniPress Ltd. Stauffenbergstr. 7. D-51379 Leverkusen Opladen, Germany 28347 Ridgebrook. Farmington Hills, MI 48334. USA www.budrich-unipress.eu Jacket illustration by disegno, Wuppertal, Germany – www.disenjo.de Printed in Europe on acid-free paper by paper & tinta, Warszaw 5 Contents Acknowledgements .......................................................................................9 Preface .........................................................................................................11 1. Introduction .....................................................................................13 2. Graduate Employment in the Era of Mass Higher Education and European Harmonisation ........................................................21 2.1 The Legacy of Mass Higher Education .............................................21 2.2 The Changing Nature of Labour Markets ..........................................23 2.3 Graduate Employment in Western Europe ........................................26 3. The Institutional Embeddedness of Graduate Employment – An Analytical Framework ..............................................................31 3.1 The Institutional Embeddedness of Life Course Transitions .............31 3.1.1 The Matching Problem as Life Course Transition Process ...............33 3.1.2 The Institutional Embeddedness of Life Courses ..............................35 3.1.3 Institutional Effects and the Transition from Higher Education to Work ..............................................................................................38 3.2 The Institutional Determinants of Graduate Employment .................39 3.2.1 The Importance of Institutions for Explaining Vocational Training to Work Transitions ............................................................40 3.2.2 The Institutional Determinants of the Transition from Higher Education to Work .............................................................................43 3.2.3 Political Economies, Institutional Complementarities and the Transition Process .............................................................................64 3.2.4 Assessing Continuity and Change in Different Political Economies .........................................................................................67 3.3 Embedding the Transition from Higher Education to Work into Varieties of Capitalism ......................................................................70 6 4. Analysing Graduate Employment Trajectories – The Methodological Approach .......................................................73 4.1 Motives and Means of Cross-National Comparisons ........................73 4.2 The Country Cases: Germany and Britain .........................................77 4.3 Quantitative Data Analysis of Individual Transitions from Higher Education to Work .............................................................................79 4.3.1 The Data Sources: German Socio-Economic Panel and two British Cohort Studies .......................................................................79 4.3.2 The Method: Event History Analysis ................................................85 4.4 Operationalisations of Concepts and Variables .................................88 5. Higher Education and Labour Market Institutions in Germany and Britain ..................................................................99 5.1 Stratification in Higher Education: Vertical and Horizontal Institutional Differentiation .............................................................100 5.1.1 The Road to Mass Higher Education ...............................................100 5.1.2 Institutional Differentiation and the Stratification of Higher Education ........................................................................107 5.1.3 Pathways into Higher Education .....................................................120 5.1.4 Stratification of German and British Higher Education: Diverse Trends .................................................................................123 5.2 Occupational Specificity: Generalist and Specialist Curricular Orientations ....................................................................125 5.2.1 Historical Frames: Professional vs. Liberal Higher Education ........126 5.2.2 General Objectives of Higher Education Institutions ......................129 5.2.3 Occupation-specific Training of Different Types of Degrees and Fields of Study ..........................................................................133 5.2.4 Professional and Liberal Orientations in German and British Higher Education .............................................................................139 5.3 Graduate Labour Market Segmentation: Between Professional and Public Spheres ..........................................................................140 5.3.1 Higher Education and the Social Closure of Professional Labour Markets ...............................................................................141 5.3.2 Public Services between Status Protection and New Public Management ....................................................................................146 7 5.3.3 Trends in Public Sector Employment ..............................................152 5.3.4 The Social Closure of Graduate Labour Markets in Germany and Britain .......................................................................................154 5.4 Labour Market Regulation and the Protection of (Un-)Employment ............................................................................155 5.4.1 Employment Protection Legislation and the Stability of Employment Relations .....................................................................156 5.4.2 Unemployment Protection and the Level of Skill Protection ..........160 5.4.3 Strictness and Flexibility of German and British Labour Markets ..165 6. Graduate Employment between Elitist Ideals and the Realities of Mass Higher Education .............................................167 6.1 Early Employment Patterns after Graduation ..................................168 6.2 Institutional Differentiation and the Stratification of Labour Market Returns ................................................................................172 6.2.1 Labour Market Returns of Initial Employment Positions ................175 6.2.2 Mechanisms of Stratification and the Transition to Employment ...181 6.3 Occupational Specificity and the Match between Field of Study and Occupation ................................................................................189 6.3.1 Transition to a Job in an Occupation Matching the Field of Study ....192 6.3.2 Occupational Mobility after First Placement ...................................200 6.4 The Importance of Occupational and Internal Labour Markets for Graduate Employment ...............................................................205 6.4.1 Labour Market Segments and Entry Ports after Graduation ............208 6.4.2 The Educational Determinants of Employment in Entry Port Occupations .............................................................................213 6.4.3 Mobility Within and Between Professional and Public Labour Market Segments .............................................................................218 6.5 Labour Market Regulation and the Stability of Employment Positions ..........................................................................................224 6.5.1 Stability of Initial Employment Positions after Graduation ............226 6.5.2 Occurrence and Effects of Unemployment ......................................229 6.5.3 Upward and Downward Labour Market Mobility ...........................233 14 graduate labour markets, is the decisive factor for shaping the career prospects of higher education graduates. It is an understanding of these national institutions, i.e. of their structure and ideational principles, which explains commonalities and differences in graduate employment across Europe. Accordingly, the main research question is: How do national institutions influence individual transitions from higher education to work? Existing research on the relationship between higher education and work has not yet systematically linked institutional arrangements of higher education systems and graduate labour markets with individual employment outcomes. It is crucial to fill this gap because it is a necessary precondition for doing expedient cross-national comparison of the transition from higher education to work. Furthermore, only an analysis of the institutional embeddedness of graduate careers can provide a blueprint for evaluating the changing nature of graduate employment in the era of globalisation, European harmonisation, and the rise of a knowledge-based society. In this regard, further theoretical conceptualisation of the institutional embeddedness of the transition process in combination with longitudinal analysis of individual career trajectories from higher education to work is needed and provided by this book. It thus fills a gap in the existing literature on comparative higher education research. It develops a novel theoretical approach for embedding graduate careers into a country’s wider political economy and identifies four institutional spheres that are important for the transition from higher education and work: the structure of higher education systems, the content of study, the structure of graduate labour markets, and labour market regulation. Empirically, this book conducts two in-depth case-studies of countries most different in their institutional set-up, Britain and Germany. The institutional analysis carried out for these two countries draws upon secondary literature and relevant policy documents published over a twenty year period between the early 1980s and the early 2000s. This institutional analysis is accompanied by a quantitative longitudinal study of graduate career development during the first five years after graduation in Germany and Britain to show how institutional configurations translate into career outcomes. The book makes three unique contributions to comparative research on the transition from higher education to work in Europe: First, it uses an innovative analytical framework for analysing the institutional embeddedness of career mobility processes; second, it combines an institutional analysis of higher education systems and graduate labour markets with a longitudinal analysis of the transition from higher education to work; and third, it covers a period of more than 20 years, which makes this book unique in its empirical scope. Regarding the analytical framework, it applies established theoretical concepts for vocational education and training (VET) to work transitions to the sphere of higher education. In the field of VET, many studies have 15 investigated the impact of various forms of education and labour market institutions on individual career development. It has become commonplace to distinguish between two basic types of institutional contexts related to the wider political economy. One the one hand, there is a more occupationalised one operating in countries like Germany, Austria or Denmark, where young people are predominantly trained within a standardised apprenticeship system and from thereon experience rather smooth trajectories into the labour market. On the other hand, a more organisation-based system is said to exist in Anglo-Saxon countries such as Britain or the US, where the transition process appears to be less structured by the education system, resulting in more protracted transition phases with stop-gap-job patterns and higher risks of unemployment (Allmendinger 1989a; Kogan and Müller 2003; Marsden 1999; Maurice et al. 1986; Müller and Gangl 2003; Shavit and Müller 1998). It is an important goal of this book to demonstrate that, for the transition from higher education to work, the effects of different political economies are as pronounced as for vocational education and training. Second, this book combines a thorough institutional analysis of both, higher education and labour market institutions with a longitudinal statistical study of individual career mobility. Particularly since the 1990s there has been a rising interest in evaluating labour market outcomes of higher education graduates in a comparative perspective (see for example Altbach and McGill Peterson 1999; Brennan et al. 1996a; Teichler 1996; 1999). Two strands of research can be distinguished. One focuses on micro level developments and analyses labour market careers of higher education graduates by comparing employment outcomes in different countries (Jahr et al. 2003; Kivinen and Nurmi 2003; Lindberg 2007; Schomburg 2007; Schomburg and Teichler 2006; Teichler 2000; 2002b; 2007a). These studies are important for demonstrating similarities as well as differences of graduate employment across countries. However, they lack a consistent institutional analysis which helps to integrate and interpret their micro-level findings. The comparative analysis of higher education policy and policy change is the focus of the second strand of research. Since the 1970s, scholars have been concerned with higher education and its function for the labour market and society as a whole (see for example Archer 1972; Ben-David 1977). In the 1990s, comparative research on higher education systems as well as on policy reforms became a “hot topic” (see for example Altbach and McGill Peterson 1999; Bleiklie and Byrkejeflot 2002; Clark 1995; Enders 2004; Huisman et al. 2001; Martens et al. 2007; Scott, P. 1995; Teichler 2007b). While these studies give a profound account of the institutional development of higher education systems and how it varies across countries, a systematic connection with individual data on career mobility has not yet been made. This book integrates both strands of research and adds a third, the study of labour markets, which has so far been neglected in comparative research on 16 graduate employment. This book introduces well-established concepts of comparative labour markets research, such as labour market segmentation and labour market regulation, into the specific realm of graduate labour markets. The central goal of this book, therefore, is to take into account higher education and labour market institutions as explanatory points of reference. Besides applying a novel analytical approach and analysing higher education and labour market institutions in combination with individual career mobility, this book is unique in its empirical scope of both institutional and career analysis, covering a period of more than 20 years. It is necessary to study a longer period of observation than has been done in previous research in order to show how institutional configurations develop and translate into labour market outcomes. This book is written against the background of the changes in higher education currently evolving across Europe. In the course of the Bologna Process, degree structures are being reformed and quality assurance and accreditation systems introduced in order to make the European higher education area the most competitive in the world. Yet, these reforms build on national systems that vary strongly in terms of higher education philosophy, structure, and graduate labour markets. It is too early to draw final conclusions about whether the Bologna Process will indeed serve the harmonisation and convergence of higher education systems and, accordingly, graduate career prospects. But in order to understand the changes it introduces not only to higher education systems, but also to the very nature of graduate careers itself, we have to go one step back and ask how national institutions influenced graduate career developments in different countries prior to Bologna. Especially since the beginning of the 1980s, the institutional variation among higher education systems and graduate labour markets became very pronounced across Western Europe. Regarding the organisation of higher education systems, countries began to develop their own political solutions to meet the challenges of mass higher education (Teichler 2007b). Some countries such as such as Britain undertook major reforms and even reversed the dual structure of universities and technical colleges by forming a unitary system. Others, such as Germany, proved to be very resistant to any attempt of re-organisation and only allowed for marginal modifications of higher education structures (Eurydice 2000). At the same time, labour markets became more diverse during these decades, in particular regarding the (de- )regulation of sectors and occupations that have traditionally been important labour market destinations for higher education graduates. For example, public service sector employment was subject to scrutiny in the wake of a “New Public Management” (Hood, Christohper 1995). Britain went furthest in its attempt to strip away previous privileges, which shielded public servants from the disciplines of the market and its attendant job insecurity, 17 while German public service still provides a high degree of job security (Hesse 2003; Hood, Christopher 2003). But professional services, too, historically marked by a high degree of social closure from external markets, have been confronted with deregulation, intensified competition, cost containment and more service-conscious clients (Lane et al. 2000). This book demonstrates that the diversification of higher education structures and the re-organisation and deregulation of graduate labour markets during the 1980s and 1990s are important factors for explaining cross-national differences of career mobility patterns among higher education graduates. In pursuit of these three tasks, the book is structured in the following way. Following this introduction, chapter 2 sets out the historical and institutional developments of higher education systems and labour markets which have led to the current situation of graduate employment in Western Europe. These macro-level developments are contrasted with actual labour market outcomes of higher education graduates in different countries in order to show where further research is needed and how this book is going to accomplish this task. Chapter 3 develops the theoretical framework for analysing the transition from higher education to work in a comparative perspective. Based on the theoretical concepts of VET research and the Varieties of Capitalism approach (Hall and Soskice 2001), it is argued that country-specific career patterns depend to a large extent on the institutional complementarities between higher education systems and graduate labour markets. The main hypothesis of this book states that country-specific linkages between higher education and the world of work create distinct transition regimes for individuals. These regimes follow a specific path-dependent logic that is intimately related to the wider political economy, and has been relatively stable over time. Four institutional spheres and their likely effect on graduate employment are explored in more detail: the structure of higher education systems (stratification), the content of study (occupational specificity), the structure of graduate labour market (segmentation), and finally, labour market flexibility (regulation). By applying the theoretical concepts related to these institutional spheres to higher education-to-work-transitions, it becomes possible to formulate precise hypotheses regarding the extent to which institutions influence graduate career mobility. Chapter 4 gives an account of the methodological approach and the methods used. The empirical analysis comprises an in-depth analysis of Germany and Britain by conducting, in a first step, an institutional analysis of higher education systems and graduate labour markets based on legal and policy documents as well as on secondary literature on policy development in the respective institutional spheres. In a second step, a longitudinal analysis of graduate labour market careers during the first five years after graduation 18 is conducted to examine whether and how institutional variations translate into career outcomes. This is done by estimating event-history models based on the German Socio-economic Panel (SOEP) and two British cohort studies, namely the National Child Development Study (NCDS) and the 1970 British Cohort Study (BCS70). Chapter 5 examines the historical development as well as the institutional set-up of higher education systems and graduate labour markets during the 1980s and 1990s in order to look for national configurations within Germany and Britain. To analyse the stratification of higher education, it gives an account of the legal basis and structural set-up of British and German higher education, and explicates the educational pathways that are necessary preconditions for entry into higher education. In order to account for varying degrees of occupation-specific training provided by higher education systems, it sketches in broad terms the different philosophies and principles that underpin higher education in both countries and examines the training of professionals, engineers and managers in more detail. The analysis of the structure of graduate labour markets focuses on the institutional organisation of labour market segments that have traditionally been associated with graduate employment, the professions and the public service. A final institutional aspect to be analysed for graduate careers is the degree of labour market regulation, as it relates to employment and unemployment protection. Chapter 6 investigates labour market careers of higher education graduates during the first five years after graduation. Each sub-section will shortly review empirical results that have been found for (vocational) education in general and for higher education in particular in order to embed the findings of this study into the broader framework of country-specific transition patterns. In particular, this chapter shows how the stratification of different higher education systems influences the status of initial employment positions and the role that higher education expansion plays in this regard. Furthermore, various forms of occupation-specific training and their importance for the match between higher education credentials and early labour market outcomes as well as for occupational mobility are studied. To assess the impact of different labour market structures, the transition to and mobility between professional and public labour market segments are examined. This is closely related to the analysis of more or less flexible labour market regulation and its impact on the stability of career paths. In the concluding Chapter 7, the empirical findings are summarised and re-analysed with special reference to the wider institutional framework and the Varieties of Capitalism approach. From the analysis conducted in the previous chapters, it follows that national institutions indeed matter; they have a strong influence on the development of graduate careers. In Germany, the transition from higher education to work follows a smooth path, while in Britain it is more comparable to a long and winding road. Based on these 19 findings, the concluding chapter will construct a common point of reference by developing a theoretical explanation for the overall transition patterns in both countries. It argues that country-specific institutional complementarities between higher education systems and graduate labour markets make it possible to provide varying degrees of skill and status protection. Due to the higher degree of skill and status protection prevalent in CMEs, the connection between higher education and the graduate labour market tends to be more “decommodified” and sheltered from market competition as compared to LMEs. This guarantees a smooth transition process, while a lower degree of skill and status protection is associated with more turbulent labour market entry patterns. The development of specific institutional complementarities between higher education systems and labour markets of different political economies leads to a more general theoretical conceptionalisation of the transition process in Western Europe. Built on its theoretical impetus and empirical evidence, this book sheds light on the factors that shape the working careers of graduates. It also contributes to the further conceptualisation of the linkage between higher education and work. In particular, it demonstrates that the skills provided by higher education and required by the labour market are not universal; rather, their demand and supply follows a national institutional logic. The skills of higher education graduates required by the labour market are shaped by specific institutional structures and historical trajectories that cannot be easily transferred from one country to the next. In the end, cross-national differences in the relationship between higher education and labour markets will therefore have to be closely monitored against the background of the Bologna Process, which aims to harmonise the structures of higher education systems across Europe. Such a perspective neglects, however, the existence of path dependencies of higher education systems and graduate labour markets. What this indicates is that national higher education reform projects initiated by the Bologna Process are likely to follow a tortuous route to convergence. This book aims to gain a better understanding of some of the obstacles that stand in its way. 21 2. Graduate Employment in the Era of Mass Higher Education and European Harmonisation 2.1 The Legacy of Mass Higher Education Historically, the remarkable changes in higher education began as early as in the first half of the 20th century when Western European universities changed from elite education to mass education (Teichler 1988). Despite existing differences between national higher education systems, countries all over Western Europe followed a similar path of development for several decades after World War II. Initially, higher education expansion was driven from below. Ongoing processes of democratisation, which enabled an ever larger number of citizens to participate in society, lead to an increasing demand in education participation among all social groups (Klemm 2000; Müller 1998). Soon more and more pupils acquired upper secondary credentials, the prerequisite for entry into higher education, and the demand for attending universities rose considerably. Given the relatively low number of university institutions in all European countries at that time, education policies had to respond to the rising numbers of students by expanding higher education institutions accordingly. The expansion of higher education accelerated in most Western European countries during the 1950s and 1960s and was marked by the optimistic belief that an increase in the number of students was the necessary and unavoidable precondition for economic, technological and social progress. One major driving force was enabling students from lower socioeconomic backgrounds to achieve higher levels of education in order to reduce social inequalities (Krais 1996; Müller 1998). Another key aspect was the increasing complexity of modern societies and economies, which required a more highly trained workforce. Overall, the shift from elite to mass higher education meant that new forms of institutional organisation and management needed to be found that could effectively and efficiently cater to more students, more diverse students, and new demands of the labour market. It was soon discovered, however, that traditional university systems, originally designed to educate the elites of society, could not fulfil their societal function anymore. As a response, major university reforms were introduced all over Western Europe during the 1960s and 1970s, which basically led to a stronger differentiation among institutions mainly through the introduction of technical colleges (Teichler 2002a). In some countries such as Germany, the higher education system responded to the new 22 challenges by establishing Fachhochschulen to respond to the labour market’s “manpower requirements” by providing education with a high degree of applied knowledge transfer and occupational training. Other nations such as Britain rather pursued the goal of increasing educational equality, polytechnics were introduced to allow more students from lower socioeconomic backgrounds to acquire higher education. By the beginning of the 1980s, however, the expansion euphoria had levelled off in a lot of countries since none of the proclaimed goals had actually been achieved. What was more, new problems resulting from mass higher education began to emerge. Despite all reform efforts, a student’s socioeconomic background continues to be a strongly influential factor determining educational achievement and labour market opportunities in many countries (Blossfeld and Shavit 1993; Krais 1996; Müller 1998). In addition, funding of higher education institutions is more and more seen as one of the major problems resulting from higher education expansion (Jongbloed 2004), leading to crowded lecture halls, rising student-teacher ratios, poorly equipped facilities, and, consequently, to a deterioration of overall study conditions. During this phase of disillusionment, countries began to search for more flexible solutions to the problems caused by mass higher education, and higher education policies in different countries started to diverge. Western European countries began to develop their own political solutions, and country-specific differences in the organisation of higher education became more pronounced (Teichler 1988). In cross-national comparison, differences were particularly strong regarding the speed and the extent of reforms. Some countries, such as Germany, proved to be very resistant to any attempt of reorganisation and only allowed for marginal modifications of university structures and institutional procedures. Others, such as Britain, undertook major changes and even reversed the dual structure of universities and technical colleges by forming a unitary system (Eurydice 2000). It was not until the end of the 20th century that European countries again sought to find common answers to their problems. In 1999, 29 European governments signed a declaration in Bologna, which proposed the creation of a European area of higher education by 2010. The Bologna Declaration represents the attempt to formulate a common European answer to common European problems in higher education. This common higher education is designed to facilitate student mobility, enhance the transparency and the recognition of qualification and quality and establish a European dimension in higher education (CRE 2000; Keller 2003). The process originated from the recognition that, in spite of their valuable differences, European higher education systems were facing common internal and external challenges related to the growth and diversification of higher education, such as the financing of university systems, the promotion of students’ and researchers’ 23 international mobility, the provision of equal access to education, the governing capacity of the state, and the changing importance of the curricular contents for the job market (Sporn 1999). The dynamics of this process continues to evolve, and the consequences for graduate careers are only starting to emerge. Optimists believe that higher education systems across Europe are eventually going to converge, which as a consequence will make career prospects more similar across countries and thereby increase the employability of higher education graduates. The effects of previous path-dependent developments on gradate careers shall not be underestimated, however. However, the present study makes use of these path dependencies and concentrates specifically on the two decades prior to Bologna in order to see how national higher education institutions and the legacy of mass higher education shape the transition from higher education to work in different countries. 2.2 The Changing Nature of Labour Markets Higher education expansion is only one of the major challenges that graduate employment is facing today. Other problems stem from changes in the labour market itself. Due to increasing rates of unemployment over past decades even among those sections of the population with higher education degrees, the social consequences of educational expansion with regard to the labour market have gained more and more attention. First of all, there were widespread concerns that large numbers of graduates might have completed their higher education in vain, either because they would end up in unemployment or would obtain low-status positions, where their competencies were not needed (Büchel et al. 2003; Müller 1998; Teichler 2002b). Even though the upgrading of participation in education was paralleled by an upgrading of labour market structures, the latter proceeded at a slower pace. Therefore, the fear of an overqualified “academic proletariat” has been prevailing since the mid-1970s and has shifted the focus of politicians, practitioners and experts to the question of labour market success of higher education graduates (Teichler 1999). Occasionally, this concern about an oversupply of graduates had been expressed earlier, but it gained momentum after the “oil-shock” in 1973, when unemployment began to grow substantially in market-oriented economies (Teichler 1999: 29). Since then, not only the risk of overqualification but also the threat of unemployment itself began to create problems for the employment chances of higher education graduates. Between 1980 and the mid-1990s, most countries in Western Europe had experienced their highest rates of unemployment since the Second World War. The persistence 30 cross-sectional data analysed at one particular point in time. They can be criticised for being too static and not taking into account the duration of transition processes and the time-dependent variations of initial labour market outcomes. On the basis of cross-sectional data it is not possible to depict the complexity of mobility patterns associated with the transition from higher education to work. If quantitative methods are used, the focus therefore should be on pathways, on life course trajectories rather than on labour market outcomes at a single point in time, to overcome a static and sometimes misleading approach for studying labour market entries. Consequently, another major purpose of this study is to analyse the dynamics of the transition from higher education to work by applying more appropriate statistical techniques. In summary, previous research on graduate employment often fails to develop a coherent explanatory framework, to take explicitly into account the structure graduate labour markets, and to adopt of a longitudinal life course perspective on labour market careers. 31 3. The Institutional Embeddedness of Graduate Employment – An Analytical Framework Finding a job after graduation is an important step in every graduate’s life. This chapter will set up the theoretical framework that explains how early career processes are shaped by national institutions. The overall hypothesis is that transition patterns from higher education to work differ systematically due to country-specific institutional linkages between higher education systems and labour markets. At first sight, labour market entry seems to be merely influenced by a student’s educational attainment, for example the type of degree obtained, the subject studied, or the type of institution attended. These characteristics seem to be essential prerequisites for any form of labour market outcomes, such as type of occupation, unemployment experiences, length of job search, or number of job shifts taking place after graduation. However, it will be argued that these individual transition processes are the result of a country’s institutional framework. National higher education systems and graduate labour markets are linked in such a way that they produce specific patterns of labour market entry and career mobility. In order to develop a comprehensive theoretical framework for studying the transition from higher education to work, it is necessary to consider concepts of education and labour market theory on the micro and the macro level. The following chapter provides an overview of the most prominent theories in these research fields. The aim of this outline is neither to provide a comprehensive summary of all existing theories nor to weigh or test them against each other, but rather to explore whether their most distinctive features are suitable for comparing labour market careers of higher education graduates in different countries. But before moving to the specific realm of education and labour market theories, it is crucial to reflect upon general assumptions of how individual life courses are shaped by national institutions. 3.1 The Institutional Embeddedness of Life Course Transitions At the micro level, the general theoretical problem of successfully assigning job seekers to jobs and thus of analysing the transition from higher education to work is often referred to as a matching problem. Economic job-matching 32 theory (Jovanovic 1979; Sattinger 1993) stresses that a good labour market match is two-sided since it not only results from an employee’s adequate education and experience but also depends on job characteristics and employer preferences. For a successful assignment to come about, employer and job applicant have to make a positive rational decision1: the employer to hire, the applicant to accept the job conditions. But since both have to make their decisions under conditions of imperfect information, job mismatches, i.e. a low degree of fit between required and acquired skills are likely to occur. Career mobility takes place because employers and employees are interested in finding an optimal match and accordingly try to re-adjust the assignment until this has been achieved. Consequently, from the perspective of job-matching theory, initial employment positions after higher education and early career mobility are the outcome of such a two-sided matching process. This micro-economic approach will be taken as a starting point for the analytical framework of the transition from higher education to work. However, it lacks two central aspects: the principle of time and the importance of institutions, which are both vital for understanding how early career mobility patterns come about. Institutions in particular are also important for explaining cross-national differences, since decision making processes not only depend on specific job requirements, preferences and recourses of the actors involved, but also on the institutional environment which affects the range and sequence of alternatives of the choice-agenda (North 1990). Country-specific variations of the institutional settings will accordingly give rise to different matching processes and create particular mobility patterns. The life course approach offers suitable concepts in this regard and will therefore be adopted to specify and enrich the theoretical model of the matching process.2 1 In this work Boudon’s (2003) premises of rational choice theory – individualism, understanding and rationality – are adopted to outline the understanding of human behaviour. 1) Any social phenomenon is the effect of individual decisions, actions and attitudes (individualism). 2) An action can be understood (understanding). 3) Any action is caused by reasons in the mind of individuals (rationality). It is assumed that actors act purposively following a set of preferences and priorities that they seek to realise in a satisfying way. This is done under bounded rationality, meaning that they act on uncertain expectations, restricted information and partial knowledge (Boudon 2003). This concept of rationality goes far beyond the mere instrumental notion of rationality, proposed by Rational Choice Theory (Coleman 1994). It rather takes into account that any social action generally depends on beliefs. Therefore both preferences and priorities as well as individual actions and choices have to be explained in terms of their meaning for the actors. They are rational not because of rational means-ends calculations, but because the actors perceive the effect of their reasoning to be rational. 2 In other words, this study analyses the institutional embeddedness of life courses, conceptualising the matching problem as a life course transition. Other perspectives on the higher education to work link, which will not be taken into consideration, might include an economic point of view on market returns to human capital (Becker, G. S. 1964; Becker, G. 33 3.1.1 The Matching Problem as Life Course Transition Process In contrast to the timeless realm of abstract social theories such as neoeconomic approaches or functionalism, life course approaches stress the importance of social institutions and time for the development of human lives. Life course research analyses the social pathways of human lives in their historical time and place, from childhood to old age, by considering how these pathways are influenced at the micro level by the course of individual action and ageing, and at the macro level by the importance of historical and geographic contexts. In pursuit of models of the life course that reflect historical and biographical time, a number of useful concepts have been developed (see Elder et al. 2003: 8-15). Each provides a way of thinking about the way lives are socially organised by showing that the time structure of individual lives is by no means only the outcome of individual choice and decision making processes but rather shaped by societal forces. A life course trajectory can be described as a “pathway defined by the ageing process or movement across the age structure” (Elder 1985: 31). In modern societies trajectories are the social pathways of education and work, family and residences that are followed by individuals and groups through society. These pathways are shaped by historical forces and are often structured by social institutions. Trajectories are themselves made up of transitions, which are changes in state or role. Examples of transitions include leaving parental home, becoming a parent, or – as it is the case in this study – entering the labour market after having finished education. The linkage between a single transition and a complex trajectory is a sequence, defined as any life course movement that includes at least two transitions between states (Sackmann and Wingens 2003: 96). The time between transitions is known as duration. Long durations enhance behavioural stability through acquired obligations and vested interests (Elder 1994). Regarding the transition from education to work and the associated matching problem, life course research has already identified several particularities associated with this particular stage in life. First of all, labour market entry after finishing education is rather a process than one single change of state or one single match (Allmendinger 1989a; b; Kerckhoff 2001). As a consequence, the transition from higher education to work is better characterised as a transition period with several transitions occurring simultaneously or sequentially. Such sequences of transitions after full-time education might include obtaining a part-time job, being unemployed for S. and Chiswick 1966; Mincer 1970; Schultz 1961), a functionalist analysis of mass higher education in modern societies (Archer 1972; Clark 1995), the spread of best practice higher education models for the labour market in a world society (Meyer 1977; Meyer et al. 1992; Ramirez and Meyer 1980), or a conflict theoretical point of view on labour market inequalities produced by the higher education system (Collins 1979). 34 some time, returning to education, obtaining a full-time job, and then being on parental leave. Transitions in early life may also have lifelong implications for shaping later events, experiences and transitions. Initial labour market experiences after leaving full-time education have a particularly long-lasting impact on career development. Furthermore, the timing of transitions is important for developmental consequences. The same events or experiences may affect individuals in different ways, depending on when they occur in the life course. All these concepts – transitions, trajectories, sequences – are closely related to the process of ageing, which has become another primary theoretical vehicle in life course research. Ageing not only indicates that a person is becoming older, but also that specific states, roles and experiences are normatively associated with a particular age in society (Kohli 1985). Since age is highly collinear with labour force experience, it is difficult to separate the effects of age, job experience and other state durations. There is, however, the possibility of examining the effect of age norms for career development. It might be that those who enter the labour market or are promoted at relatively young ages are perceived as those who are especially able and likely to achieve more. Moreover, it is expected in general that the rate of job shifting slows down with age or labour force experience due to a declining discrepancy between current and potential positions, or due to the development of skills particular to a given job or employer (Rosenfeld 1992). Particularly in different national contexts, the age norms of finishing higher education and entering the labour market might strongly differ and have to be taken into account. The duration of the matching process is therefore of utmost importance for analysing early labour market careers of higher education graduates. It is essential to pay attention to the length of early labour market spells, to take into account the timing of transition events during the life course, and to focus on sequences of transitions occurring during a longer transition period after finishing education instead of analysing only a single point in time. In addition to these duration aspects of the transition process, the quality of early labour market outcomes is certainly important as well. A very basic criterion in this regard is that transitions after graduation can occur between different life phases, such as the transition from higher education to work, from work to unemployment, or from work to parental leave. But transitions can also take the form of changes between different labour market states, such as being employed full-time, then part-time, and then out of the labour force. Furthermore, transitions between different kinds of jobs with particular employment characteristics, such as those transitions between various types of occupations or between high and low status positions, are essential for characterising the quality of the matching process. In this regard, mobility studies have pointed to the importance of job shifts for career development. 35 While a job is a particular kind of work with a particular employer, a career can be defined as a sequence of jobs, usually with some notion of progress or at least coherence (Rosenfeld 1992: 40). Job shifts can be considered as the building blocks of individuals’ careers and thus as important transitions during career development. With the help of the information about the duration and the quality of the matching process taking place after graduation, it becomes possible to investigate sequences of transitions in the beginning of graduate careers that, taken together, constitute specific transition patterns from higher education to work. This conceptualisation of the matching process will be used as explanandum for the cross-national analysis of graduate career trajectories. However, the concepts discussed so far still remain at the individual level and pay attention to the conduct of individual lives based on rational decision making, while they neglect their institutional embeddedness. In this regard, life course research also stresses also that certain macro level phenomena, namely institutions and historical circumstance, are decisive for structuring the nature of individual life courses. 3.1.2 The Institutional Embeddedness of Life Courses Ryder (1965) has shown that historical changes often have different implications for people of different age, because these age differences become manifest in different experiences and recourses when confronted with new situations and consequently lead to different adoption processes to new conditions. In general, a cohort is a group of people who have shared the same critical experience during the same interval of time, for example in the most common sense the year of birth, but also the year of labour market entry or of retirement (Alwin and McCammon 2003; Heinz 1996; Ryder 1965). Thus, cohorts link age and historical time by showing how people of different age bring different resources to situations and consequently adapt differently to social conditions. When historical change differentiates the lives of successive cohorts, it generates a cohort effect. For the transition from education to work, it has been shown that as a consequence of educational expansion, younger cohorts face different employment prospects than older ones (Falk et al. 2000; Hillmert 2001). In addition, history can take the form of a period effect when the impact of social change is relatively uniform across several cohorts. Both period and cohort effects constitute evidence of historical influences, which always have to be taken into account when explaining cross-national similarities and differences of transition patterns. These multiple meanings of history brought time and temporality of the macro level to life course thinking. However, it is not only the principle of time that explains the social embeddedness of lives, but also the principle of 36 place or societal institutions. The institutionalisation of life courses refers to the process by which normative, legal or organisational rules define the social and temporal organisation of human lives. This institutionalisation either refers to stages or states in life, which can be formally or informally structured like education, employment or partnership, or to life course events and transitions like leaving school, entry into and exits from labour contracts, or ages of pension entitlements (Kohli 1985; Weymann 1989; 2003b). Life course research remains very vague in its definition of institutions and often refers to institutions of the welfare state as major determinants for the organisation of modern life courses (Mayer and Müller 1989). Explicit theoretical accounts of institutions, however, are missing. Therefore, this study adopts the comprehensive definition of institutions given by Scott (2001) who sees institutions as social structures that have attained a high degree of resilience and are composed of regulative, normative and culturalcognitive elements that provide stability and meaning to social life. Regulative elements refer to a society’s formal and informal rules such as legislation or traditions. Normative elements define roles and expectations for its members. And cultural-cognitive elements structure the knowledge of how appropriate behaviour in a society looks like (Scott, R. W. 2001). According to Scott (2001), institutions operate at multiple levels of jurisdiction, from the world system to localised interpersonal relationships, and by definition connote stability while also being subject to change (Scott, R. W. 2001: 47f). In this conception, institutions are multifaceted, durable social structures, made up of material resources, social activities, and symbolic elements. They have developed historically in a time and placespecific manner and are thus path dependent. Despite being relatively stable over time, they are also subject to change, both because of the impact of the broader institutional environment in which they are embedded and because of the impact of the aggregation of lives that follow these pathways. This draws attention to a dynamic analysis of institutional development taking into account the historical processes of institutional evolution, stability and change.3 This societal macro level of institutions enables and constrains individual life course conduct by setting a regulative, normative and cognitive framework for individual preferences, capacities and expectations. Institutions can affect individual behaviour by giving information about the legitimacy of social practices (Scott, R. W. 2001), by providing actors with greater or lesser degrees of certainty about the appropriateness of behaviour 3 A definition of institutional path dependency and explicit mechanisms generating institutional stability and change remain underdeveloped in life course research. The dynamics of institutional development will therefore be examined more thoroughly within the Varieties of Capitalism approach (Hall and Soskice 2001), which offers more theoretical considerations in this regard (see p. 54ff of this study). 37 of other actors (March and Olson 1989), by affecting the range and sequence of alternatives of the choice-agenda (North 1990), or by supplying actors with moral or cognitive templates for interpreting actions (Meyer and Rowan 1977; Zucker 1977). Actors are thus choosing between different possibilities by referring to an institutional environment that gives meaning and thus perceived (bounded) rationality to their actions. The major focus of this study will be on regulatory institutions, the formal and informal rules guiding job search and job placement activities after graduation. Regulatory institutions that are especially relevant for shaping individual life course transitions in modern times are national life course policies, such as education, labour market, unemployment, family, or retirement policies. There have been different cross-national attempts to link theories of welfare regimes and political economies to the life course approach, and thereby develop a comparative model of nations from a life course perspective (Leisering and Leibfried 1999; Mayer and Müller 1989; Weymann 2003a). According to these studies, country-specific life course policies lead to different life course regimes, and, consequently, to different patterns of transition within life course stages. The term life course regime indicates that the life course in one country has some overall logic that reflects the institutional structure (Leisering and Leibfried 1999: 14). Country-specific political regulations provide institutional support for the conduct of life by giving structure and helping to stabilise expectations, transitions, and trajectories. For the transition from higher education to work, both education and labour market policies play an important role in life course politics, for example by providing age entry and exit norms of the higher education system, by defining the available range of educational credentials, by establishing rules for the distribution of labour market outcomes and rewards, or by providing assistance in the case of unemployment. Family policies, too, are important institutions for shaping labour market careers since they influence the way work and family can be combined, an essential aspect especially for the life courses of women. In addition to regulatory institutions, normative and cognitive institutions are also taken into account when explaining transition patterns from higher education to work. Normative institutions, for example, influence the expectations regarding the skills and knowledge a higher education graduate is confronted with when applying for a job. Cognitive institutional elements influence the range of jobs students consider appropriate for their type of higher education and accordingly influence their labour market behaviour. Institutions are not free-floating, however, they must manifest themselves in one way or another to be applicable to individual lives. Some authors consider organisations as such manifestations of societal institutions (Meyer and Rowan 1977; Scott, R. W. 2001), which mediate between institutions and life courses. However, contrary to the point of view of 38 sociological neo-institutionalism, which considers organisations to be institutions themselves and thus prevents an analytical separation of the two concepts, organisations in this study are regarded as distinct from institutions in terms of their main principle, the way they operate, the factors that influence their establishment and dissolution, etc. North (1990) explicitly differentiates institutions from organisations: While institutions are defined as rules and the related enforcement characteristics of these rules within a given social system, organisations are groups of individuals bound together by a common purpose (North 1990: 5). Organisations just as individuals follow the rules given by institutions. In the field of higher education, the regulative institutions of higher education policies define the structural set-up of university organisations, their financial structure, or their personnel recruitment practices. Furthermore, societal expectations define what kind of values and normative functions universities have to fulfil in society. Consequently, for explaining cross-national similarities and differences in the transition from higher education to work, relevant organisations such as different types of universities organisations, or various forms of employment organisations have to be taken into account. 3.1.3 Institutional Effects and the Transition from Higher Education to Work From applying the concepts of life course research to the labour market entry of higher education graduates in different countries it follows that the overall aim of this study is to analyse the duration and the quality of the matching process and to explain how these two are influenced by the institutional setup of higher education systems and labour markets. Rather than focussing on a single point in time, it is important to follow graduates from the end of higher education throughout their early labour market careers to analyse the sequences of life course states, labour market states, and/or job shifts they are experiencing. The timing of transition processes has to be taken into account as well, since transitions have different implications for later career development, depending on when they happened during the life course. For explaining the similarities and differences observed with regard to the labour market entry of higher education graduates in different countries, national higher education and labour market institutions serve as important blueprints against which individual transition processes may be studied cross-nationally. This does not mean that these institutions determine individual career development completely, but that they enhance people’s ability to act by providing general preconditions for individual behaviour required for achieving individual projects in a country-specific way. In this regard, an analysis of the historical evolution of the institutional structure is 39 pivotal in establishing which national rules, norms and cognitive scripts are related to the labour market entry of higher education graduates. They influence what both students and employers perceive as “adequate” graduate employment, constitute the perceived arena of choice and transition options, and define the value of university certificates as well as higher education in general. These overall effects of the national institutional environment can be called institutional effects since they form a national frame of reference. Apart from these more general institutional effects stemming from the institutional set-up of higher education systems and graduate labour markets, historical events can also have an impact on the transition from higher education to work. As a cohort effect they result in similar work trajectories of students that share a common historical experience, such as being born at the same time, or, what seems even more important, having graduated at the same time. If history takes the form of a period effect, such as German reunification, it influences a country’s population as a whole. Only by analysing the various time dependent and independent impacts of institutions on life courses one is able to gain deeper insights in the formation of national transition regimes. But which institutional features of higher education systems and labour markets are important constituents of a national transition regime? In order to establish a more distinct picture of the social mechanisms underlying the transition from higher education to work, concepts of existing education and labour market theories will be introduced in the following sections and applied to explain patterns of graduate career mobility. 3.2 The Institutional Determinants of Graduate Employment Institutions of higher education and labour markets shape matching processes as they affect both worker and employer action in the labour market. One can define higher education as any kind of post-secondary education which is either largely theoretically based and intended to provide sufficient qualifications for gaining entry into advanced research programmes (for example history, philosophy, mathematics, etc.) or professions with high skill requirements (for example medicine, dentistry, pharmacy, etc.), or which is rather practically oriented, designed for participants to acquire the practical skills and know-how needed for employment in a particular occupation (for example architecture, engineering, management studies, etc.) (UNESCO 1997). Higher education institutions define the design of education programmes, the range of achievable credentials, the structural set-up, governance, and financing of higher education organisations, or the societal norms and cognitive scripts associated with this kind of education. Labour 46 Various types of institutions or degrees can project both a horizontal and a vertical differentiation within the higher education system, depending on whether they are hierarchically structured (such as BA and MA) or not. Different fields of study normally pertain to the same hierarchical level and therefore only differentiate the system horizontally. Institutional reputation, in contrast, by definition creates a hierarchy among institutions and is thus exclusively related to vertical differentiation. Both forms of differentiation might stratify the labour force; however, only a high degree of vertical differentiation is assumed to stratify labour market outcomes hierarchically and as such is associated with a high degree of stratification. As a consequence, occupational rewards of initial employment positions in vertically differentiated systems should be more unequal in status. H1.1: The higher the vertical differentiation of a higher education system, the stronger differ the (status) returns to different higher education tracks. The other dimension of stratification – the proportion of a birth cohort that attains the maximum number of school years provided by the education system – does not have to be adapted for higher education graduates, since it directly refers to the number of students obtaining a higher education degree. Based on this dimension, stratification can be taken as a good indicator of higher education expansion. It is important to note, however, that this dimension does not refer to the structure of higher education, but rather to the set-up of secondary education and its sorting function for entry into higher education. This sorting in secondary schooling acts as an important filtering mechanism for the process of higher education expansion since it restricts the number of students with entry certificates. The basic premise of this mechanism is that in secondary education systems with a low level of stratification, less sorting takes place through secondary schooling, which, as a consequence leads to a higher proportion of higher education graduates. In highly stratified systems, on the other hand, the main sorting of pupils into different tracks occurs already in secondary education, which subsequently yields lower graduation rates in higher education. In some systems, therefore, it is easier to obtain the highest available degree than in others based on the structure of secondary education. Accordingly, the degree of a country’s higher education expansion is not only the result of individual choice (more students willing or demanding to attend higher education) and higher education structures (more available higher education institutions), but also depends on the selectivity of secondary education. Building on this mechanism, Müller and Shavit (1998) argue that increases in the proportion of higher education graduates tend to decrease returns to the educational level below, but also overall status outcomes of higher education graduates (Müller and Shavit 1998). Their theoretical 47 explanation is based on Thurow’s labour queue model (Thurow 1975), which assumes that most skills necessary for job performance are obtained on the job. Educational qualifications are not valued by employers for the skills the represent per se, but for the indirect information (signals) they provide, for example, about a candidate’s intelligence (trainability), work habits, or disciplines. Due to the importance of these indicators, employers establish a hierarchical queue among applicants based on their educational credentials, which places higher education graduates ahead of other qualification holders. If such a job-queue is at work, there is a built-in incentive for young people to acquire more education qualifications in order to stay ahead of the queue. But as ever larger proportions of a given age cohort obtain the highest credentials, the labour market value of these credentials declines (Müller 1998; Müller and Shavit 1998). Thus, secondary education systems with a low level of stratification tend to produce an excessive supply of higher education graduates, thereby lowering the value of these credentials in the labour market. By contrast, in systems with highly stratified secondary education, the value of higher education credentials is preserved because only a lower percentage of a birth cohort is able to attain a higher education degree. This indicates that in (secondary) education systems with a low level of stratification, the labour market returns of higher education graduates in general should be lower than in stratified systems, where a smaller proportion of higher education graduates are still able to outperform the rest of the labour force. At the same time, the time needed to find a job matching the level of education should be much shorter in stratified than in unstratified systems. H1.2: The higher the stratification of secondary education and the lower the proportion of a birth cohort that obtains a higher education degree, the higher the overall labour market returns of initial employment positions after graduation. H1.3: The higher the stratification of secondary education and the lower the proportion of a birth cohort that obtains a higher education degree, the less time it takes graduates to obtain favourable labour market positions. In summary, an analysis of higher education stratification operating in different political economies will show how the flow of students is channelled to initial employment -positions. At the level of higher education, a high degree of stratification leads to strong status differences between students with different higher education credentials and strongly stratifies labour market returns accordingly. Vertically differentiated higher education systems in particular are supposed to allocate individuals to positions of matching status within the differentiated occupational structure and to 48 decrease the duration of the job search to achieve a good match. Higher education systems that are highly stratified by type of institutions, degrees, institutional reputation, or fields of study should be associated with trackspecific timing and matching effects. The stratification of secondary education and the degree of higher education expansion more generally affects the comparative advantage of higher education graduates as compared to qualifiers from other educational levels. If secondary education is highly stratified and the proportion of higher education graduates remains low, graduates will generally stay ahead of the queue and obtain higher-status positions as compared to secondary education systems with a low level of stratification, where credential inflation lowers the overall labour market returns of higher education graduates. Since the concept of stratification refers to the structural set-up of education systems, it is a good example of how regulatory institutions are likely to influence the transition from higher education to work. The concept of occupational specificity is less concerned with institutional structures, but focuses more strongly on the content taught at higher education institutions. Therefore, it serves as an example of how norms and historical ideas have an impact on career development.er education tracks. Occupational Specificity and the Nature of Training According to the concept of occupational specificity, higher education systems may vary in the extent to which they offer curricula that are designed to prepare students for particular occupations and award credentials that are occupationally specific. Occupations can be defined as groups of individuals who perform similar activities, have similar skills and competencies, as well as similar amounts of social power deriving from their occupational position within the social division of labour (Kalleberg and Sorensen 1979: 361). For both employers and employees, occupations serve as a meaningful internal structure of the labour market, facilitating the acknowledgement of status transfer as well as the process of matching qualifications to skill requirements. However, the way qualifications correspond to certain occupations and accordingly the match between contents learned in the education system and job requirements differ cross-nationally. Comparative studies on VET systems have shown that only in systems with a high degree of occupational specificity, a close match between vocational training and occupation is achieved immediately after leaving education, while in systems with a stronger focus on general skills the training for specific occupations does not take place within the education system, but has to be attained after entering the labour market (Gangl 2000a; 49 2001; Kerckhoff 2001; Müller and Shavit 1998).7 The theoretical argument holds that educational resources are not beneficial for individual career outcomes per se, but rather that the structure of education systems as providing either general training or more specific occupational training has an effect on individual labour market entry via the different linkages established between educational and employment systems. The micro mechanism associated with occupation specific training is related to employers’ relative reliance on formal qualifications vs. labour market experience as a reliable measure of individual skills during recruitment and allocation processes. To the extent that educational credentials do little to convey job applicants’ potential capabilities for a particular occupation, employers will be more likely to assess individual skills either from their past work records or from training on the job. In turn, if training systems provide qualifications that are meaningful indicators of skills for a particular occupation, employers should be more likely to use this inexpensive signal of individual capabilities and hire job applicants on the basis of their credentials. Interestingly, the concept of occupational specificity helps to overcome one of the major criticisms of neo-economic models of the matching process, namely that employers possess perfect information about the applicants’ skills (Kalleberg and Sorensen 1979). The assumption that a high degree of occupation-specific training can influence employers’ expectations about what has been learned in the education system implies, first of all, that education is indeed able to transfer specific knowledge and skills to students, and second, that employers possess perfect knowledge about the relationship between productivity and specific educational certificates. Accordingly, in occupation-specific systems, more information about productivity is available to both employers and job applicants and matching of applicants to jobs can be based merely on credentials. In systems conferring more general skills, on the other hand, employers cannot trust that a specific level of 7 Becker’s version of human capital theory explicitly draws attention to the problem of general vs. specific skill acquisition (Becker, G. S. 1964). While general human capital is useful in many firms, (firm-)specific human capital is only beneficial to the firm that has provided it through training on the job. In contrast to firm-specific training, general skills either are acquired through the formal education system or are provided by firms, if they offer general training in form of industry-specific apprenticeships, which provide skills that are useful for the whole industry. The main problem of general skills is the possible poaching of skilled workers by other firms. As a consequence, firms are reluctant to invest in general skills, since they are not willing to pay for skills can be used anywhere else. Instead, individuals have to pay for the acquisition of general skills themselves, by receiving no wages at all (as it is the case for general school education) or by earning wages below what could have been received elsewhere (as it is the case for apprenticeships). Firms are willing to pay for specific training since they benefit from higher productivity. At the same time no reasonable employee would invest in training that cannot be used anywhere else, especially when the rate of labour turnover is high. 50 productivity is related to certain credentials. Instead, when hiring people they have to screen individuals more thoroughly, have to train them on the job, which makes labour market mismatches more likely. The major problem of applying the concept of occupational specificity to higher education is related to empirical measurement. In the sphere of vocational training, the degree of occupational specificity depends upon the number of standardised apprenticeships designed for a specific occupation and the orientation of curricula towards the teaching of specific vocational knowledge (Kerckhoff 2001; Müller and Shavit 1998). Measurement of the occupational specificity of higher education is not as straightforward, since no such entity as an apprenticeship exists. When compared directly to vocational training, higher education is often perceived as conferring only general skills and therefore national variations of occupational specificity are considered low (Estévez-Abe et al. 2001). However, it is also very likely that on the level of higher education countries place different emphasis on particular types of skills. The analysis of the importance of occupationspecific training associated with the different dimensions of institutional differentiation should yield information in this regard. The content taught by different types of higher education institutions, degree courses, and fields of study as regards their orientation towards specific knowledge, but also the incorporation of practical training may serve as appropriate measures of occupational specificity at the higher education level. In addition, the organisation of professional training should be essential, too, since professions in particular are characterised by a highly exclusive knowledge base, systematic professional education and training, and accordingly specific fields of action relevant only for this particular occupational group (Abbott 1988; Brater and Beck 1981; Heidenreich 1999). The effect of occupation-specific training related to any of the discussed dimensions of institutional differentiation should be a smooth transition between higher education and work, where higher education credentials are the primary criterion for job placement. Qualifications with a strong focus on occupational training should result in a close match between acquired and required skills of the first job, since both employers and graduates possess enough information about the kind of education needed. As regards the duration of the matching process, occupation-specific training should also guarantee that employment in an occupation matching the higher education credential is found more quickly after graduation. In systems which exhibit only a low degree of occupational specificity, the transition process appears to be less structured by higher education credentials, while discretionary recruitment strategies, and other individual signals should play a crucial role in shaping early labour market careers. If the specialisation is low, it should not take long before graduates get their first job, since they can choose from a large variety of different occupations. However, more general certificates 51 should result in a larger variation of occupations obtained with the same type of higher education qualification and in longer transition periods until a matching job is found. H2.1: The higher the occupational specificity of a higher education credential, the closer the match between educational certificate and occupation of the first job. H2.2: The higher the occupational specificity of a higher education credential, the shorter the duration of finding an occupation matching the qualification after graduation. Apart from finding a matching job immediately after graduation, patterns of occupational mobility after higher education should also differ between higher and lower degrees of occupational specialisation. General skills are usually transferable between different occupations, while the transferability of specific skills is more limited. This indicates that students with more general skills are more likely to experience shorter initial labour market spells due to an inappropriate match. Until a matching job is found, extensive job hopping between different types of occupations should be the consequence, often in combination with some training on the job to create a firm-specific skill profile. Students trained for a specific occupation, in contrast, should – once they are in employment – keep the same occupation for a longer time as a result of the good initial match. The high occupational specialisation should decrease the number of shifts between occupations and increase the chance of stable employment in the beginning of graduate careers. H2.3: The higher the occupational specificity of higher education, the lower the likelihood of changing the type of occupation after job shifts. To sum up, the degree of occupational specificity related to a higher education credential influences both the quality of the match between qualification and first occupation and the time it takes before a matching job is found after graduation. But despite the assumed commonalities between occupational specificity of VET and higher education, there might also be an important difference. In cross-national comparisons of vocational training, a country’s whole system is classified as having a high or low degree of occupational specialisation, which as a consequence yields similar transition patterns for any student undergoing vocational training. Due to the different dimensions of institutional differentiation, occupation-specific training in the field of higher education might be associated with particular types of institutions, degrees, or fields of study rather than with a country’s higher 52 education system as a whole. Accordingly, different degrees of occupational specialisation might not only account for cross-national variations, but also produce strong intra-national differences. It remains to be seen empirically whether differences are more pronounced between countries or between different higher education credentials. Excursus: Standardisation and Political Centralisation A concept that is often used in cross-national comparisons of education systems is standardisation. It describes the degree to which institutional features of an education system meet the same standards nationwide (Allmendinger 1989a; Kerckhoff 2001; Müller and Shavit 1998). In the field of secondary education and vocational training, for example, researchers working with this concept analyse the extent to which degrees, curricula, or teacher training are standardised throughout a country and how this influences transition processes (Allmendinger 1989a; Bernardi et al. 2004; Hannan et al. 1999). Regarding the function of standardisation for labour market entry, it is argued that the setting of standards provides a valuable tool for evaluating what students have learned during their studies and what to anticipate for their performance in work life. Thus, the concept of standardisation influences individual job perspectives mainly with respect to future employers’ expectations because in standardised systems employers can assume the presence of certain contents already learned in the education system. As a consequence, they can trust the information given by certain standardised certificates and do not have to screen and/or train individuals entering the labour force. A higher degree of standardisation is therefore assumed to lead to a smooth transition process between education and work, making it easy to find a first job soon after leaving education. Furthermore, repeated job shifts should not be necessary for obtaining stable employment. If educational processes and related qualifications are not standardised, they cannot be taken as accurate indicators of educational achievement. Students educated in unstandardised systems should have longer search phases and more job shifts in the beginning of their career. In the field of VET to work transitions in particular, the concept of standardisation is often closely related to the occupational specificity of an education system, because only if occupation-specific credentials are standardised at the national level (e.g. as the German dual system) can they provide reliable information to employers about whether certain occupationspecific skills have been learned. In general, however, literature dealing with these concepts remains ambivalent whether standardisation can be considered an explanatory concept by itself in cross-national comparisons. While Allmendinger (1989a) and Kerckhoff (2001) consider it as a separate institutional dimension, Müller and Shavit (1998) find that the effect of standardisation disappears once stratification and occupational specificity are 53 controlled for (Müller and Shavit 1998: 38). Allmendinger in her more recent work also relates standardisation specifically to vocational training (Allmendinger and Hinz 1998). In addition, empirical studies on the occupational specificity of vocational training systems do not separate standardisation analytically from the degree of occupational specificity (Gangl 2000b; 2001; Scherer 2001; 2005). However, this does not mean that the concept for cross-national comparisons should be abandoned completely, but rather that we should be more explicit about the institutional dimensions that are standardised throughout a country. If the content of education credentials follows particular occupation-specific standards, then standardisation clearly is part of the more encompassing concept of occupational specificity, since only standardised occupation-specific credentials lead to a close match between education and initial employment positions. But degrees, curricula, or examinations can also be standardised at the national level without necessarily conferring occupation-specific skills. This might particularly be the case at more general levels of education, such as secondary schooling or higher education. As regards secondary education, studies have shown that both examination level and examination grades are very important for employers’ decisions in highly standardised systems, while no significant effects are evident in less standardised systems (Breen et al. 1995; Rosenbaum and Kariya 1991) In the case of higher education, research has shown that there are considerable historical and cross-national variations in the degree to which higher education systems are integrated with the political structure, yielding more or less standardised institutional elements. In highly centralised higher education systems, governments intervene universally and thereby define set nationwide standards regarding admission to higher education, the organisation of curricula, or the examination of students. In contrast, in more decentralised systems these institutional features vary more strongly between different higher education institutions (Archer 1972: 7). For the transition from higher education to work this implies that in highly standardised and centralised systems employers should find it rather easy to choose between applicants with the same degree, since they simply choose the applicant with the better results. In less standardised systems neither the certificate nor the achieved result carries enough information to allow employers to make a meaningful decision between candidates. Accordingly, the degree of standardisation of a higher education system can be considered to be a result of the centralisation or decentralisation of higher education governance within a country, if particular standards are predominantly established by national governments. Since the country-cases examined in this study do not reveal significant variation in the degree of standardisation of higher education credentials and since, furthermore, data on exam results are not 54 available in the used data-sets, the concept of standardisation cannot be tested empirically and will therefore be left out in the following analysis. Nevertheless, standardisation should be included as a valuable concept for explaining cross-national similarities and differences in the transition from higher education to work, if other country-cases are compared and better data is available. Cross-national Differences in the Institutional Set-up of Graduate Labour Markets The discussion of the institutional embeddedness of transition processes has so far been centred around the organisation of higher education systems and their likely impact on graduate career development. The theoretical arguments put forward have shown how country-specific transition patterns might come into place due to vertical and horizontal institutional differentiation and the occupation-specific contents of various higher education credentials. However, a mere focus on higher education institutions would fail to consider the ways in which the world of work imposes its own requirements on the higher education system. After the initial matching problem is solved, i.e. obtaining the first job after graduation, the major problems are how career development is organised and how formal degrees and certificates correspond to the hierarchies and divisions of labour within and between firms. Specific forms of higher education and training need a structure of career mobility into which they can be integrated, and graduates need potential career rewards if they are to be persuaded to undertake the effort, if not the forgone earnings, and expense, of higher education. After graduates enter the labour market, it is therefore important to study how further career development is influenced by variations in the institutional set-up of graduate labour markets. From an institutional perspective, graduate labour markets are not only equilibria of demand and supply of skills just as the ‘market metaphor’ of human capital theories implies, but a combination of institutions and processes which determine the pricing, buying, selling and allocation of labour (Kalleberg and Sorensen 1979). Two different forms of labour market dynamics might be possible for explaining transition patterns from higher education to work. Graduate labour markets might either function in relative independence from the higher education system, mainly following their its own logic, or alternatively, they may strongly build on the institutional structure of the higher education system. There are various micro theories that explain how and why early career mobility is likely to proceed. At the most basic level, one may argue that career mobility is a consequence of an increasing accumulation of human capital. As soon as additional forms of human capital are acquired, employees are likely to seek promotion and 55 employers are willing to grant it. In addition to formal certificates, general work experience or on-the-job training may signal such an increase in human capital and accordingly lead to promotion (Becker, G. S. 1962; Mincer 1970; Schultz 1961). Job matching theory, on the other hand, would predict that labour market mobility is the result of an initial job mismatch rather than of human capital growth (Jovanovic 1979; Sattinger 1993). Since employers and employees are dissatisfied with a job allocation, in which skills do not meet job requirements, both are prone to induce job shifts in order to achieve a more satisfying match. Particularly in the case of overqualification, where the level of education acquired exceeds the level of education required for a particular job, upward mobility after initial job placement is likely to occur (Büchel et al. 2003). Supporters of job vacancy theories loosen this tight connection between educational attainment and career mobility. They consider job vacancies an important influence on individual career development, since labour market entry and mobility depend on the availability of vacant positions. The basic idea is that a person’s move to a new job creates a vacancy to be filled and therefore creates a new mobility opportunity for another person. Matching and promotion probabilities are considered to be dependent on vacancy creation, on the distribution and source of vacancies (exits, new jobs), the number and nature of competitors, and managerial decisions about how to fill vacant positions (Kalleberg and Sorensen 1979; Rosenfeld 1992; Sorensen 1977). The main contrast between job vacancy approaches and human capital theories can be found in the assumptions they make about educational influences on career mobility. While human capital theories argue that (income) mobility occurs only in accordance with increases of human capital, job vacancy approaches claim that educational resources do not change over time in an organisation and thus have no necessary relationship to the opening of vacancies and resulting mobility processes. The main function of educational resources in this perspective is to determine the best possible position a person can fill. Again, overqualification may occur since at least some people will enter an organisation in positions not matching their qualifications. However, the longer they stay with the organisation, the more they can make use of vacancies, and the smaller the gap will be between their potential and their current labour market position (Rosenfeld 1992). The discussion of higher education institutions and their impact on transition patterns has shown that they can mainly account for the match between higher education credential and initial employment position. However, they fail to explain individual job shifts and, accordingly, patterns of job mobility. The concepts of labour market segmentation and labour market regulation are well-established theoretical approaches to explain cross-national variations of career mobility; yet, they have not been applied to the field of higher education exclusively. Both theoretical strands are able 62 as legally binding obligations, where contracting parties can seek juridical enforcement in case they disagree about certain conditions. The degree of employment protection legislation influences how easy it is to hire and fire employees, how quickly job offers can be adjusted to new conditions, how willing employees are to accept these new terms, and how strongly the form of employment contract can be individualised. In highly deregulated systems employers’ are free to hire and fire in order to adapt their firms’ workforce more easily to changes in product markets, technologies or corporate restructuring. At the same time, a high degree of freedom to adjust job offers to new conditions allows an immediate reaction to changes in employment circumstances and in the overall economic situation. Furthermore, if employers have the freedom to influence the form of employment contract, it is easier for them to respond to the high uncertainty of pace and direction of economic change, such as the outdatedness of certain skills. Strict employment protection legislation, on the other hand, essentially serves to stabilise existing employment relationships by restricting employers’ rights to terminate the contract at will and by restricting the use of short-term or temporary contracts. Since the degree of employment protection legislation affects all workers and employees to the same extent, no adaptation of this concept to graduate labour markets is necessary. Labour market protection influences both the ease with which graduates can be laid off as well as the ease with which unemployed workers are reabsorbed back into employment. Therefore, one can expect that stricter EPL is likely to lower job mobility rates among higher education graduates since it restrains employer-initiated labour turnover and makes it less likely that graduates will be laid off even during economic downturns. Consequently, once graduates have found their first job, employment relationships will be more stable and fewer job shifts will occur. H4.1: The stricter labour market regulation, the more stable initial employment positions after graduation. In particular, unemployment protection legislation influences the way unemployment impacts on further career development in relation to occupational mobility. A high degree of unemployment protection normally goes hand in hand with protection from income reduction due to unemployment and reduces the uncertainty over wage levels throughout one’s labour market career (Estévez-Abe et al. 2001; Gangl 2002b). High replacement ratios and longer benefit duration permit the unemployed enough time to find another job that matches their skills, especially if they are permitted to turn down jobs that are outside their core competencies. This ensures that their reemployment will generate the same occupational 63 outcomes as before, simultaneously reducing occupational mobility. Therefore, generous systems of unemployment benefits may prolong the period of adjustment, encouraging the unemployed to engage in a long period of job search instead of settling for one of the jobs on offer. In systems with low unemployment protection, employers are free to adjust job offers to new economic circumstances and unemployed job seekers are likely to adjust their expectations downwrds, in line with the new set of job offers. H4.2: The stricter labour market regulation, the lower the occupational mobility occurring after phases of unemployment. Overall, the direct effect of labour market regulation on job mobility patterns should be to subdue involuntary occupational mobility and to stabilise employment relationships. But taken together, the degree of labour market regulation is also said to influence status mobility patterns (Gangl 2002b). By lowering labour market turn-over and by insuring that occupational positions are maintained even after phases of unemployment, strict labour market protection reduces the associated risk of downward mobility. However, this rather positive outcome of a strict regulatory framework also has its downside. Just as it reduces the risk of downward mobility, it is likely to reduce upward mobility, too. In this regard, Gangl (2002b) has argued that a high degree of labour market regulation has an indirect effect on mobility by lowering overall vacancy levels in the labour market and thus shortening mobility chains. Upon entering the labour market, graduates continue to acquire human capital, since job experience will continuously add to their productive capabilities. In less regulated labour markets, this skill enhancement should result in upward mobility, be it firm-internal or in a different company, to achieve better matches of individual skills and job requirements. However, since stricter legislation is likely to shorten mobility chains, graduates are less likely to experience upward mobility even after human capital increases in highly regulated labour markets. While being effective in protecting individuals’ current positions, a high degree of labour market regulation might be unfavourable to career mobility dynamics because any reduction in overall labour market turnover implies reduced opportunities for upward mobility. H4.3: The stricter labour market regulation, the lower status mobility among higher education graduates, both upwards and downwards. Overall, labour market regulation should influence the stability and the quality of career mobility during the transition from higher education to work in external labour markets. In less regulated labour markets, graduates are likely to experience unstable career mobility with high degrees of 64 occupational mobility after spells of unemployment as well as upward and downward mobility. In other words, rigid labour markets, while providing a higher guarantee for long-term employment, maintenance of occupational outcomes after unemployment, and less downward mobility, should also be likely to inhibit upward mobility. 3.2.3 Political Economies, Institutional Complementarities and the Transition Process The discussion of the various institutional spheres has demonstrated how stratification, (standardisation), occupational specificity, labour market segmentation, and labour market regulation might influence the matching of graduate skills to jobs and as such the individual transition from higher education to work. By systematically applying these concepts to graduate career trajectories, it will be possible to explain the country-specific transition patterns graduates follow after finishing higher education. Yet, what all these approaches fail to make clear is whether and how the different institutional spheres interact with each other, and whether they are linked in any particular way. The general aim of this study is not only to obtain a large variety of more or less distinct labour market outcomes, but to identify how country-specific transition regimes are systematically produced by the institutional coupling between higher education systems and graduate labour markets. In this regard, the Varieties of Capitalism approach (Hall and Soskice 2001), popular in political sciences, offers helpful insights. Even though the theory on Varieties of Capitalism does not explicitly refer to individual career mobility, it provides a systematic approach to understanding the coupling between different institutional spheres. The application of the concepts developed in this strand of theorising can therefore help to explore the institutional embeddedness of career trajectories between higher education and work. The Importance of Institutional Complementarities for Linking Institutional Spheres Instead of focussing on the impact of institutions on individual life courses, the Varieties of Capitalism approach (see Hall and Soskice 2001 for the following) takes the point of view of individual firms, analysing how different institutional environments influence the way they solve their coordination problems in five different institutional spheres: vocational education and training, employee relations, industrial relations, inter-firm relations, and corporate governance. The main argument holds that differences in the institutional framework of the political economy generate 65 systematic differences in corporate strategies. The core distinction is drawn between two ideal types of political economies, liberal market economies (LMEs) and coordinated market economies (CMEs). In LMEs, firms coordinate their activities primarily via hierarchies and competitive market arrangements, whilst firms in CMEs depend more heavily on non-market relationships, such as networks, collaboration, and trust. The notion of institutional complementarities is central to this approach, since the presence of institutional complementarities between these five different institutional spheres reinforces the overall differences of the political economies. According to Hall and Soskice (2001), two institutions are complementary if the presence (or efficiency) of one institutional sphere increases the returns from (or efficiency of) another (Hall and Soskice 2001: 17). The institutional complementarities between VET systems and labour markets of different political economies identified by this theory are of particular interest for this study. According to Varieties of Capitalism, in the sphere of vocational education and training, firms are confronted with the problem of securing a workforce with suitable skills, while workers face the problem of deciding how much to invest in general or specific skills. Firms in CMEs, on the one hand, rely on a high-skill equilibrium with high skill protection. They employ production strategies that are based on high-quality products, which require a highly skilled labour force with substantial work autonomy, and consequently depend on a training system that provides workers with such skills. The matching problem in CMEs is solved by a publicly subsidised VET system that is supervised and coordinated by industry-wide employer associations and trade unions. This high level of skill protection ensures both that the training fits the firms’ skill requirement and that free-riding on the training provided by others is limited. Firm behaviour in LMEs, on the other hand, is based on a low-skill equilibrium with low skill protection. The production strategy employed aims at standardised mass products and relies on highly fluid labour markets with high job turnover, which quickly have to respond to changes in consumer demand. This encourages employees to invest in general skills transferable across firms and industries, and training is normally provided by formal education. At the same time, firms are reluctant to invest in standardised apprenticeship schemes imparting occupation-specific skills since they have no guarantee that employees will stay long enough to make training costs pay off, or that other firms will not simply poach their apprentices. Due to this lack of skill protection, companies in CMEs do a substantial amount of inhouse training which provides employees with firm-specific skills that cannot easily be transferred from one firm to the next. Even though Hall and Soskice do not consider individual labour market transitions and their relation to different political economies, one could argue that country-specific transition regimes from VET to work are a result of 66 institutional complementarities between education systems and labour markets. The specific form of institutional complementarities in CMEs, which is based on high skill protection, is the prerequisite of the tight coupling between vocational training systems and labour markets. In this regard the organisation of a standardised and industry-wide apprenticeship system, where collaboration between employer organisations, trade unions and national governments guarantees that quality and quantity of vocational skills closely match labour market demands, can be considered an institutional basis for a smooth transition from vocational education to employment. The absence of a standardised vocational education and training system in LMEs in combination with market-led labour market matching and training-on-the-job are indicators of a low skill protection regime, which makes skill investments more risky and results in a loose coupling of systems. Labour market integration of young people is more turbulent and flexible, as it depends on market forces and discretionary employer recruitment. Table 2 summarises this line of argument by bringing together the theoretical concepts of the Varieties of Capitalism thesis and vocational education to work transition research. Table 2: The institutional determinants of VET transition patterns in political economies CMEs LMEs Secondary schooling (Allmendinger 1989) high degree of stratification in secondary schooling low degree of stratification in secondary schooling + + Vocational education (Kerckhoff 2001, Müller and Shavit 1998) importance of occupationspecific skills importance of general skills + + Labour market segmentation (Marsden 1990, 1999) occupational, sectoral labour market segmentation (OLM) internal, enterprise labour market segmentation (ILM) + + Labour market regulation (Estévez-Abe et al. 2001, Gangl 2002b) strict labour market regulation and legislation weak labour market regulation and legislation ↓ ↓ Education-occupation link (Maurice et al. 1986) qualificational space, tightly coupled organisational space, loosely coupled ↓ ↓ Institutional complementarities (Hall/Soskice 2001, Estévez-Abe et al. 2001) specific skill equilibrium, high skill protection general skill equilibrium, low skill protection ↓ ↓ Patterns of labour market entry smooth, late, integrated labour market entry turbulent, early, stop-gap labour market entry 67 In brief, CMEs provide more institutional support for the strategic interactions required to realise the value of co-specific assets, primarily in the form of industry-specific training of standardised apprenticeship systems. The more fluid markets of LMEs provide economic actors with greater opportunities to move their resources around in search of higher returns, encouraging them to acquire switchable assets, such as general skills. Since in Varieties of Capitalism research, political economies are conceptualised as institutional complementarities between different economic spheres, higher education systems can also be understood as a complementary in national production systems. This should result in specific institutional complementarities between higher education and labour markets, which are likely to influence the labour market careers of higher education graduates in a similar manner and constitute the basis of national transition regimes. Therefore, the nature of institutional complementarities in different political economies can be considered the major cause of similarities and differences in graduate employment across Europe. In the existing literature on the transition from higher education to work, this institutional linkage has not been made explicit yet. The following empirical investigation explicitly aims at establishing the institutional complementarities between higher education and labour markets in LMEs and CMEs. It remains to be seen whether these institutional complementarities between higher education and work follow the same logic as the complementarities between vocational education and work and, in turn, entail analogous transition patterns. 3.2.4 Assessing Continuity and Change in Different Political Economies A last comparative aspect of the transition from higher education to work is related to continuity and change. As indicated in the introduction, changes at the macro level are predominantly related to higher education expansion and the changing structure of labour markets. At the micro level, the ongoing ageing process throughout the life course and its connection to historical time, mainly in the form of cohort and period effects, is a key influence on individual labour market entry and career mobility. The major purpose of this last section is to link these two levels of analysis in a dynamic way to develop a theoretical model that takes into account the importance of time for institutional and individual development. This should lead to a more encompassing perspective on observed similarities and differences of graduate employment in different political economies. At the institutional level, issues of continuity and change have been mainly addressed by arguments of institutional convergence and path dependency. Regarding convergence, for example, it has been argued in the 68 field of research on higher education institutions that over time, national university systems have been becoming more and more alike due to the experience of common difficulties. Theorists proclaiming convergence start with the observation that massive changes in the institutional environment of higher education, such as the higher education expansion of the last decades in combination with structural changes stemming from globalisation, Europeanization and the rise of a knowledge-based society, create problems that countries have to respond to (Altbach and McGill Peterson 1999; Sporn 1999). All higher education systems have been subject to the same, or similar, pressures – to widen access and increase participation, to operate more efficiently, and to develop more relevant teaching and research programmes. As a result, countries should pursue similar reform strategies, and their higher education systems should become more and more alike (Scott, D. K. 2002). Based on the analysis of institutional development, similarities of graduate employment across Europe could, thus, be explained by converging structures of higher education systems. The Bologna Declaration on the European Space of Higher Education (CRE 2000) seems to be the logical consequence of this line of argument: The introduction of the same degree structure in all signatory countries is expected to serve convergence and standardisation of institutions as well as transition regimes. However, the convergence argument neglects the country-specific historical development of institutions (Ebbinghaus 2005; Hall and Soskice 2001; Mayer 2001; Streeck and Thelen 2005). The notion of path dependence offers a promising possibility for a more general explanation in this regard. Two major reasons are often given to justify the existence of path dependency. The first argument for the existence of path dependency claims that once a country or region has started down a certain path, it is likely to stay on it in the future, since initial choices are not easily reversed, and paths cannot be left without large costs (see Ebbinghaus 2005 for a discussion). This notion of path dependency refers to those particular sequences which have self-reinforcing properties (Pierson 2004). Institutions are path dependent because of increasing returns encouraging processes where “each step along a particular path produces consequences which make a path more attractive for the next round” (Pierson 2004: 5). Through these processes, institutions provide resistance to pressure that might otherwise force actors off a particular path. As a result of these increasing returns, inefficient institutions may well persist over time. Thus the path dependency of institutional arrangements tends to be highly stable over time and does not allow for abrupt changes. The second argument for the existence of path dependence refers to institutional complementarities. Hall and Soskice (2001) claim that although firms attempt to sustain or restore the forms or coordination on which the broader political economy has been built, these efforts may entail changes to 69 existing institutions or practices in the economy. Institutional complementarities are assumed to play an important, but ambiguous role in these processes of adjustment (see Amable 2000: 656-660). The basic hypothesis of institutional complementarities in this regard is that they generate disincentives to radical changes, since firms attempt to preserve arrangements in one sphere of the economy in order to protect synergies with institutions that are of value to them. Thus, several institutions taken together reinforce each other so that they form a coherent and stable but not everlasting structure, while another combination of these institutions would be unstable. The coherence of a national political economy – defined as a set of interrelated national institutions – is thus the expression of complementarities between specific institutional arrangements. A given set of institutions can either reinforce each other over time (which explains the emergence of a particular structure of institutions), or they tend to weaken the coherence of the structure over time (which explains why institutional structures, national models, or political economies are not eternal). Thus, in contrast to the first argument for the existence of path dependency, which stresses the spontaneous evolution of an institution and its subsequent long-term entrenchment, the existence of path dependent institutional complementarities can explain both institutional stability and change. The complementary connection of different areas of the economy can provide institutional stability through coherence over time. This also means that institutional changes in one sphere might have global effects for the whole system, since every complementary relationship with other institutions will be undermined and the stability of the whole system could become at risk (Hall and Soskice 2001). In this regard, the notion of a hierarchy among institutions is crucial. Institutional hierarchy points to the relative importance of one or a few institutions for the coherence and dynamics of the institutional architecture of the whole system. The circumstance in which such a hierarchy emerges also depends on history. For explaining the dynamics of institutional change, the existence of an institutional hierarchy stresses that only the transformation of a hierarchical superior institution affects the evolution of others by destabilising the coherence of the whole institutional architecture and by threatening the positive complementarities existing between institutions (Amable 2000). Since country-specific transition regimes can be considered a result of institutional complementarities between higher education systems and labour markets, this line of reasoning is of immediate relevance to the analysis of institutional stability and change and its relation to the transition from higher education to work. Following from the argument above, the overall assumption is that institutional complementarities between higher education and the world of work are rather stable over time. This stability leads to a coherent transition regime from higher education to work. For the field of 70 vocational training, the notion of a coherent national system can be used as an explanation for the question why the introduction of a more standardised vocational training system in liberal market economies has failed so far. Since it is not complementary to the national system of production and the structure of its labour market, employers do not have incentives to make use of a standardised vocational training. This points to the fact that, on the one hand, changes in one institutional sphere – be it the higher education system or the graduate labour market – do not necessarily lead to a change of the whole transition regime. These changes might influence individual transition patterns in the form of period or cohort effects, but do not imply that the coherence of the whole system is in danger. Thus, in contrast to the convergence hypothesis, the convergence of life course regimes does not unavoidably follow from similar national strategies of reforming higher education. On the other hand, the existence of institutional hierarchies also raises the prospect that institutional reform in one sphere of the economy could snowball into changes in other spheres as well. In the field of vocational training, the labour market in LMEs seems to be the hierarchically superior institution, while it is rather the standardised vocational training system in CMEs that has stronger effects. In general, it is claimed that in negotiated economies adjustment processes to external shocks are often slower than in economies coordinated primarily by the market (Hall and Soskice 2001). In brief, for analysing continuity and change in the transition from higher education to work it is important to reflect on the coherence of the national system as well as to assess the hierarchy of institutions. But even if single institutional features of higher education systems or graduate labour markets change, the overall institutional complementarities and the resulting transition regimes should be rather stable over time. 3.3 Embedding the Transition from Higher Education to Work into Varieties of Capitalism The point of departure in the previous sections was the claim that an empirical investigation of graduate employment has to include an analysis of the institutional set-up of educational systems and labour markets, as these institutions shape labour market career patterns by affecting both employee and employer action on the labour market. Based on the theoretical concepts of vocational education to work transitions and the Varieties of Capitalism literature, it can now be argued that country-specific transition patterns from higher education to work depend to a large extent on the coupling between higher education systems and labour markets, which is a result of 71 institutional complementarities of different political economies. The question is whether the institutional complementarities between higher education systems and labour markets follow a logic similar to that for vocational training and labour markets, which, as a consequence, entail Varieties of Capitalism specific patterns for the transition from higher education to work. Similar patterns of the transition process could follow from the fact that both vocational training and higher education are situated in the same political economy and should therefore exhibit similar complementarities to the labour market. As a consequence, the transition from higher education to work, too, should be less institutionalised in LMEs, which means that human capital has to be acquired and invested according to market rules. The pathways into the labour-market are likely to be much more flexible, while in CMEs all steps seem to depend on occupation-specific certificates, which are unavoidable prerequisites for progression. In order to get a clearer picture of what the institutional complementarities between higher education and work might look like in different political economies, it seems useful to study the impact of institutional elements on individual labour market entry and career mobility. The application of the theoretical concepts of stratification, occupational specificity, labour market segmentation, and labour market regulation to the transition from higher education to work are thought to give first insights of how cross-national institutional variations matter. While stratification should mainly impact status differences between various types of higher education institutions, degrees, or fields of study, occupational specificity is likely to influence the match between higher education credentials and initial employment positions as well as the occupational mobility occurring thereafter. Labour market segmentation points to the fact that higher education credentials should form the prerequisite for entry into public or professional segments of the labour market, where further career development proceeds in a way that is rather sheltered from the market. A high degree of labour market regulation is supposed to result in early labour market experiences of high stability and continuity, preventing downward status mobility, but also inhibiting upward movements. By analysing how these institutional features impact individual life courses in different countries, it should be possible to establish a broader picture of cross-national similarities and differences in graduate employment. If the different institutional spheres have effects similar to those for VET to work transitions, institutional complementarities between higher education and labour markets are also likely to follow an analogous logic. Therefore, in a first step the impact of each institutional sphere has to be studied separately. In a second step, the combination of these institutional effects will allow for constructing institutional complementarities between higher education systems and labour markets to be related to the wider political economy. So far, it remains an open question what the institutional 78 difficult. The choice was made to study higher education and labour market institutions in association with the respective graduation patterns until 2000. There are two reasons for choosing this particular year. First, eschewing the very recent past reduces the danger of overestimating recent changes relative to earlier changes. The other reason to end in 2000 is more pragmatic. In 1999, with the Bologna Process, a major new reform agenda was initiated within the European Union that aimed at changing the institutional structure of universities across Europe substantially. The Bologna Declaration represents the attempt to formulate a common European answer to common European problems of higher education, particularly by introducing common degree systems. With that step, the signatory countries agreed on major reforms of their higher education systems to be completed by 2010. While earlier changes of higher education institutions were at least partly a result of choices within the respective countries, the Bologna Declaration meant that in all higher education institutions, an externally imposed new structure was implemented. To prevent this turning point in higher education policies from interfering with developments of previous decades, both institutional data and the individual graduation year end in 2000, at a point before a rising number of graduates trained within the reformed systems entered the labour market. Thus, the country-specific paths of higher education systems and graduate labour markets end with the beginning of Bologna. At the institutional level, the time period from 1980 to 2000 poses no problems in terms of comparability between Germany and Britain. In both countries, governments with a conservative agenda (the British Conservatives with Margaret Thatcher and John Major as prime ministers and the Christian Democrats with Helmut Kohl as chancellor) dominated most of the period politically. Both countries are highly developed capitalist industrial societies, displaying substantial similarities in terms of industrial development, endowment with natural resources, IO membership (EU, OECD, NATO, etc.) as well as in political and cultural heritage. Especially their common membership in the European Union has not only served to homogenise many political and economic activities, but has also presented the two nations with similar economic and social concerns and opportunities, which have increasingly become comparable constraints. The introduction made clear that policy changes in both countries were the result of similar economic problems and similar political ideologies. Yet, a closer investigation of higher education systems and labour markets in Germany and Britain in the empirical chapters of this book will reveal that by the end of the 1990s, relatively little convergence towards a common European type of both higher education systems and graduate labour markets had occurred. Moreover, even superficially common structural arrangements hide enduringly distinctive national ways of thinking 79 about and organising the transition from higher education to work. The unique character of German and British higher education and labour market systems is reflected in the wider political economy and is a result of its historical development and present state. The institutional analysis, therefore, aims at determining these pathdependent differences. Based on the theoretical concepts of stratification, occupational specificity, labour market segmentation and labour market regulation, the institutional infrastructure of higher education systems and graduate labour markets of each country are analysed with a particular focus on developments in the 1980s and 1990s. The comparison will be based on secondary analysis of academic literature on the relevant topics as well as on official documents (for example legislative bills, statutes and opinion papers) and national statistics released by the respective political institutions, such as employment and graduation statistics. Emphasis will be placed on a broad description of general national differences rather than on a detailed analysis of particularities within each institutional sphere. This might lead in part to an oversimplification of institutional arrangements by levelling out variations within each country. At the same time, such an approach is necessary to illustrate cross-national diversity more clearly. Empirical findings will be linked to theoretical considerations to formulate more country-specific hypotheses about individual transition patterns. 4.3 Quantitative Data Analysis of Individual Transitions from Higher Education to Work Building on the findings of the institutional analysis, a quantitative longitudinal data analysis of individual transitions from higher education to work is conducted in order to relate the institutional infrastructure to life course outcomes. For studying higher education graduates’ career development at the individual level, this study examines their life courses empirically, following them from graduation to their first professional placement and the subsequent course of their career development. In the following sections, the data-sets to be used will be described in detail and the modelling strategy to be applied will be explained. 4.3.1 The Data Sources: German Socio-Economic Panel and two British Cohort Studies The analysis of the individual transition from higher education to work is carried out based on the German Socioeconomic Panel (SOEP) and two 80 British cohort studies, namely the National Child Development Study (NCDS) and the British Cohort Study 1970 (BCS70). These studies are large-scale, multi-purpose surveys that collect information on different areas of social life, such as education and employment, household formation and dissolution, or fertility. The SOEP is a longitudinal survey of private households in Germany, issued by the SOEP Group of the German Institute for Economic Research (DIW) in Berlin (Haisken-DeNew and Frick 2005). It is conducted as a panel survey, i.e. the same individuals are re-interviewed on an annual basis on a large variety of topics, including information on labour market positions, educational attainment, attitudes or family status. The longitudinal development of labour market mobility is measured in two different ways. Information on labour market states, positions and industries are gathered as panel information on a yearly basis at the time of the interview. Furthermore, labour market careers are measured retrospectively with calendar information on periods of education, employment, unemployment and periods out of the labour force, which occurred between two successive panel waves. These spell data are measured in discrete time intervals on a monthly basis for each respondent since his or her entry into the sample. Starting in 1984, the SOEP represents the residential population of the Federal Republic of Germany in 1984 including West Berlin, and later the German residential population of the former German Democratic Republic in June 1990, including East Berlin. All SOEP samples are multi-stage stratified random samples which are regionally clustered (Haisken-DeNew and Frick 2005). In the West German sample, selected foreign groups were oversampled. Furthermore, the sampling probability for the Eastern sample is bigger than the probability for the main sample in West Germany. Those different sampling probabilities were chosen to make sure that the number of cases for the different groups in the sample are large enough for analysis (Haisken-DeNew and Frick 2005). The study at hand takes account of all respondents with a higher education degree who graduated in the years 1984 – 2001 while surveyed by the SOEP, meaning that a total number of 878 graduates were included in the calculations.8 8 This means that all SOEP samples have been used. Sample A „Residents in the FRG“ covers persons in private households with a household head who does not belong to the main foreigner groups of „guest workers“. Sample B “Foreigners in the FRG” covers persons in private households with a household head from Turkey, Greece, former Yugoslavia, Spain or Italy. Both were drawn in 1983. Sample C “German Residents in the GDR” was drawn in 1990 and covers persons in private households where the household head was a GDR citizen. Sample D “Immigrants” started in 1994/95 and consists of households in which at least one household member had moved from abroad to West Germany after 1984. In 1998, a new sample E “Refreshment” of the population of private households in Germany was selected independently from the ongoing panel, following the same selection scheme also used for the selection of sample A. Sample F “Innovation” – also selected independently from all other samples – again was drawn from the population 81 Data for Britain are drawn from the National Child Development Study (NCDS) and the 1970 British Cohort Study (BCS70), which are both conducted by the British Joint Centre for Longitudinal Research9. Even though the most common British data-set used for comparisons with the SOEP is the British Household Panel Study (BHPS), it is not used for this study, since it does not contain sufficient information on higher education obtained by its respondents. The cohort studies, in contrast, provide contemporaneous information on higher education achievements as well as detailed information on early labour market careers and mobility developments (see Bynner et al. 2001). The NCDS is a panel study of all the children born in the first week of March, 1958 (n=17.634). Six follow-up surveys were carried out, when the children were of age 7, 11, 16, 23, 33 and 42 years. For the analysis at hand, data from sweeps 3 to 5 are used. The employment history in the NCDS is drawn from the retrospective responses given by individuals in the fifth wave of the survey. The BCS70 began in 1970, when data were collected about the births and families of babies born in England, Scotland, Wales and Northern Ireland in the week of April 5th- 11th, 1970 (n=16.571). To date, there have been five attempts to gather information from the full cohort, at ages 5, 10, 16, 26 and 29-30 years. For the current analysis, sweeps 3 and 5 are used, with the last one providing detailed information on the education and work history since age 16. In order to achieve a period of labour market entry that is more or less comparable with the German data-set, both cohort studies were pooled. The final data-set covers 3805 individuals graduating in the period between 1979 and 1997. Sampling Strategy and Panel Attrition Even though both data sets are conducted as panel studies, there are important differences to be taken into account when analysing them comparatively. While the SOEP is a representative sample of the German of private households in 2000, following more or less the same procedure as for sample A and E (Haisken-DeNew and Frick 2005). Despite knowing that the use of sample A-F leads to an overrepresentation of certain groups (for example foreigners in sample B or immigrants in sample D), the whole sample is chosen to have a sufficient number of cases for each of the surveyed years. Only sample G “Oversampling of High Income”, which was drawn in 2002, is excluded as the available work history of likely graduates within this sample is too short for the respective quantitative modelling. 9 The Joint Centre for Longitudinal Research comprises the Centre for Longitudinal Studies (CLS), Institute of Education, University of London; the International Centre for Health and Society (ICHS), University College Medical School, London; and the National Centre for Social Research (Natcen). Natcen was responsible for assisting in the development of instrumentation, the development of CAPI programmes, the conduct of fieldwork, initial data coding and editing, and some documentation. The British Office for National Statistics was responsible for the co-ordination of funding on behalf of various government departments. 82 population and therefore covers a broad range of age groups, dates of the NCDS and BCS70 only refer to two birth cohorts born in 1958 and 1970 respectively. Estimations based on such a cohort design represent only the cohorts included – in this respect, the parameters are cohort-specific, and their implications might be specific for these cohorts rather than a country’s whole population. Therefore, the sample design of the SOEP is, compared to the cohort design of NCDS and BCS70, clearly the more appropriate basis for estimating population parameters, and the selection of cohorts is a strategic and far reaching decision. Apart from mere practical reasons, namely the availability of sufficient information on higher education, the strongest arguments for choosing the two British cohorts are the type of analysis conducted and their comparability with the German data set. The main goal of statistical modelling in this study is not to obtain descriptive population estimates, but rather causal interrelations between variables in order to examine the relation between institutional environments and individual career trajectories. These relations should be similar for different cohorts living in the same institutional environment, and, as it has been argued in the theory chapter, should also be rather stable over time. In addition, the inclusion of both cohorts allows for investigating a similar period of labour market entry as that covered by the German sample, namely the 1980s and 1990s. This would not have been the case for the BHPS, which only covers the 1990s. Another major difficulty that arises from comparing such different sample designs lies in the differentiation between period and cohort effects. In the German sample, it is easier to include parameters that clearly distinguish both, while in Britain, it is more difficult to relate changes to any of these two effects (Scott, J. and Alwin 1998). Despite the apparent advantages of the German data set, the panel design as used by the SOEP has to solve considerable problems with population representation as well. Due to panel mortality and attrition, the sample size decreases wave by wave. This can lead to severe problems for making reliable population estimates over time. There would be no problem in the case of random non-response; however, there is little reason to expect that this is the case (Haisken-DeNew and Frick 2005; Scott, J. and Alwin 1998; Solga 2001). Hence, modelling occupational mobility in the SOEP over time might be biased since occupationally mobile respondents may be more likely to drop out of the panel due to residential mobility connected to new jobs. In addition, the multi-time measurement of information in the panel design carries the risk of changing measurements of occupational mobility, which might lead to biased comparisons over time or might even make comparisons impossible. As far as sample attrition in the cohort studies is concerned, it has been possible in both cohort studies to closely monitor the evolving rate and pattern of non-response. While it is clear that various biases have 83 emerged, these appear to be rather small and, moreover, they seem to be much the same from one study to the other. For both cohorts, sample attrition is most apparent among those with minority ethnic status, with disadvantaged class backgrounds, with teenage or single mothers, and with low levels of educational attainment (Plewis et al. 2004). One way of dealing with the problem of sample attrition is the use of sample weights. However, in the case of panel data, conventional sample weights are not appropriate. Instead, longitudinal attrition-adjusted sampling weights are needed. One important criticism regarding those weights is that they implicitly assume independent censoring, i.e. that individuals who drop out behave in the same way after their last interview as they did before (Hill 1997). To the contrary, it is quite likely that panel drop-out is related to an important life change taking place, such as a finishing university or becoming unemployed, which led to a change in the person’s behaviour. Consequently, longitudinal weights are not the best way of dealing with the problem of panel attrition.10 In addition to these more practical reasons against the use of sample weights to account for panel attrition, the modelling strategy applied in this study (see next section for more details) has an important advantage in dealing with this problem. It has been shown that weighting problems only occur for models using continuous time, while this problem does not occur in event history analysis if the models are specified for discrete time (Hartmann 1997). Hartmann’s (1997) analysis of marital stability indicates that it is possible to estimate discrete time models, such as logit models or cloglog models (Allison 1984; Jenkins 2004) without including sample weights and nevertheless obtain unbiased parameter estimates. He shows formally that the odds ratios in discrete time models are the same, whether or not sample weights are applied. Only for the estimation of standard errors sample weights have to be applied. Since all estimations made in this study make use of discrete time models, and furthermore, the major focus is on transition patterns, i.e. the direction of effects, rather than on absolute values of parameters or standard errors, it does not include sample weights in its models. Survey Design Another important difference between the SOEP and the NCDS/BCS70 is in the way of gathering information. Even though all surveys are conducted in form of a panel, the time interval between the different waves (SOEP) or 10 Applying the SOEP standard panel weights did not seem plausible, either, as these are designed to apply to the population as a whole, and not only to higher education graduates. In addition to the reasons mentioned, the construction of graduate-specific weights is very difficult due to a lack of representative data for the respective period in Germany. 84 sweeps (NCDS/BCS70) differs substantially. While, in the SOEP, panel information is collected on a yearly basis in combination with retrospective questions on education and labour market mobility of the past 12 months, sweeps of the two British cohorts cover a much longer period, lasting from four to ten years. Thus, even though NCDS and BCS70 are panel studies, their data collection closely represents those of retrospective life history studies. Both types of survey design have their distinct advantages and disadvantages. A problematic issue in terms of education and labour market mobility measures stemming from the SOEP panel design is the difficulty of capturing duration dependence. Even though event histories are gathered on a monthly basis through retrospective questions, the collected data only reflects changes in labour market states (such as changes between education, full-time employment, unemployment, or family leave), while more precise information on type of education, type of job, occupational position, or industrial sector of employment is only gathered on a yearly basis. As a consequence, it is quite complicated to rearrange the discontinuous information about status, events and transitions – given in the several waves – into a continuous flow of life histories. The most serious problem stemming from this survey design is the fact that there is not enough information about shifts between occupations and industries that may have occurred between two panel waves while being full-time employed in the retrospective calendar. As a consequence, estimates of labour market mobility based on the SOEP might be biased since the study tends to underestimate the mobility observed. At the same time, the panel design carries the risk of overestimating labour market mobility due to nonrandomly missing information on job titles and the risk of inconsistency in the frame of reference and the meanings of instructions between several panel waves (Solga 2001). In contrast to the “pure” panel design used by the SOEP, data on education and labour market mobility in the two British cohort studies is collected through retrospective life histories that go backwards from the current state to the date of the previous sweep, and sometimes even further. (The labour market history of NCDS sweep 5 conducted in 1991, for example, covers every month since January 1974, even though another sweep was conducted in 1981). According to Elder (1992), a life history is “a lifetime chronology of events and activities that typically and variably combine data records on education, work life, family, and residence (Elder 1992: 1122). Thus, instead of yearly records of mobility variables provided by the SOEP, the cohort studies allow continuous records for several life domains to be collected. The strength of this design is that for each point in time the status in the life domain as well as relevant covariates can be captured more precisely. However, since respondents are asked to recall the 85 “continuous” history of events and transitions on a monthly basis, there are restrictions for estimating unbiased parameters across time as well, namely recall error and missing information because of memory (Solga 2001). Both problems stem from the fact that remembering is a constructive process, since memory is inevitably open to a range of distortions and reinterpretations in the light of subsequent knowledge and experience (Scott, J. and Alwin 1998). Sometimes, measures of past events and transitions are unavailable both because of memory lapse and because the information is only available at the time. Furthermore, individuals can misclassify episodes or events or recall them inaccurately due to errors in their recollections of timing or dating of events. It is stated in general that the longer the recall period, the greater the concerns about the reliability of retrospective data. Regarding labour market mobility, retrospective designs involve the risk of underestimation, since recall errors may smooth out the true extent of mobility due to unreported job changes (Solga 2001). In summary, all data sets used in this study have their respective advantages and disadvantages, be they related to sampling strategy or survey design. Nevertheless, their applicability and comparability has already been proven in numerous studies. For the research problem at hand, a comparison of SOEP, NCDS and BCS70 is considered the most appropriate way of exploring the institutional embeddedness of the transition from higher education to work. 4.3.2 The Method: Event History Analysis In order to statistically model labour market entry and early career development of higher education graduates in both countries at the individual level, discrete time piecewise constant exponential models of event history analysis have been employed (Allison 1984; Blossfeld and Rohwer 1995). By means of event history analysis (also referred to as survival analysis), we may examine how the probability of an event progresses longitudinally, i.e. over time, and furthermore, how this probability varies according to relevant covariates. Cross-sectional designs (which only estimate the overall probability of an event without making reference to the duration at all) and longitudinal panel designs (which merely observe changes in labour market states between two points in time and accordingly only take the initial and the final stage of a certain period into account) are less appropriate for studying mobility processes than event history analysis. With this modelling strategy, all relevant changes of the condition during a time period as well as the timing of events can be considered. This is particularly important for analysing career trajectories, since not only the quality of the matching process matters, but also the time it takes to obtain such a matching job. 86 Although continuous-time survival analysis is frequently used in many settings (see Blossfeld and Rohwer 1995), discrete-time survival analysis is often more natural in social and behavioural science applications where time is likely to be measured discretely (Allison 1984). This study also makes use of discrete-time models instead of continuous-time models. The major reason for this strategy lies in the fact that time in the SOEP is coded discretely, namely on a monthly rather than on a more detailed basis, and only discrete time modelling makes it possible to take into account the fact that the exact exit times are not known. We only know that they fall within a certain interval of time (Jenkins 2004). A rather practical advantage is the fact that discrete-time models can easily accommodate time-varying covariates, and, as already stated above, produce unbiased estimates even if sample weights are not included in the models. The central concept for describing the timing and duration of transition processes with event history analysis is the hazard rate. According to Allison (1984), the hazard rate describes “the probability that an event will occur at a particular time to a particular individual, given that the individual is at risk at that time” (Allison 1984: 16). In the present study, the simplest form of a hazard rate is the probability of getting a first job within a particular month after graduation for those who have not yet had a job. Allison (1984) stresses that the hazard rate is an unobserved variable; however, since it controls both the timing and occurrence of events it is the fundamental dependent variable in event history models. In discrete time models, the exact point of time of the transition is unknown; instead, the interval hazard rate h(aj), i.e. the probability of exit in the interval (aj-1, aj] is defined as11 h(aj) = Pr (aj-1 < T ≤ aj | T > aj-1)12 with T defined as a discrete random variable that indicates the time period or interval when the event occurs. Another distributional representation of event time is the survivor function. The survival probability at time period j is defined as the probability of not experiencing an event (i.e. the probability of “surviving” through time period j). In discrete time, the survivor function at the time demarcating the start of the jth interval is defined as S(aj-1) = P (T > aj-1) while the value of the survivor function at the end of the jth interval is S(aj) = P (T > aj). 11 The formulas for describing discrete-time event history models are taken from Jenkins 2004. 12 The value of the discrete hazard rate differs from the continuous hazard rate in that the former is a (conditional) probability with values between zero and one, while the latter may be greater than one, since it is not a probability, but refers to the exact exit time rather than an interval (Allison 1984; Jenkins 2004). 87 A final important feature of event history analysis is the concept of right censoring, which occurs when it is unknown whether an individual experiences the relevant event because the observation period ended beforehand. This might be due to panel attrition, i.e. the individual does not take part in the survey any longer, or due to the conclusion of the observation period. Event history models can take this fact into account by calculating both the likelihood of individuals experiencing an event while at risk and or those who are right censored. As there is no single shape for the hazard function that is appropriate in all contexts, the analysis of event history data requires some model assumptions on how hazard rates vary with time. A model that does not impose too many restrictions regarding the shape of the hazard function and that, furthermore, has already proven its validity for studying education to work transitions is the piecewise constant exponential (PCE) model (Bernardi et al. 2004; Falk et al. 2000; Hillmert 2001). Its flexibility stems from the possibility of allowing hazard rates to vary between different time periods (Allison 1984; Blossfeld and Rohwer 1995; Jenkins 2004). Thus, with the discrete time PCE model any variation in the hazard between any numbers of time intervals can be estimated. Furthermore, PCE models do not require a hazard-related proportionality assumption (Jenkins 2004) that is commonly used in entirely non-parametric survival models, such as the Cox proportional hazards model. An important question in this regard is how to define the number and length of time intervals in between which the hazard is likely to vary most extensively. Previous studies on the transition from school to work have shown that differences of the hazard rate for finding a job are especially pronounced during the first year after graduation (Bernardi et al. 2004; Falk et al. 2000; Hillmert 2001). Furthermore, Jenkins (2004) suggests that the decision of whether the hazard function generally increases or decreases with survival time should be based on the data, rather than specified a priori. Taking both positions into consideration in order to avoid a random setting of time intervals, the bands of the time intervals in this study will be smaller during the first year and will be derived from the descriptive analysis of the survivor function. In order to derive an estimate of parameters describing the continuous time hazard, whilst taking into account the nature of the banded survival time data, calculations are based on the logistic hazard function, which is a straightforward way of analysing discrete time data with piecewise duration dependence (Allison 1984; Jenkins 2004). The logistic hazard model assumes that the relative odds of making a transition in month j, given survival up to end of the previous month, is summarised by an expression of the form. h(j, X) = [1 + exp (-αj – βXi(j))]-1, 94 matching occupation are supposed to be much higher in occupation-specific systems. The basic criterion for assigning occupational codes to a field of study is the assumed congruence of skills acquired through the subject and those needed in the job.16 For example: the category of sciences includes 16 This operationalisation of the congruence between fields of study and occupations is largely in line with other studies examining the match or mismatch between educational credentials and labour market outcomes. For example, Wolbers (2003) uses a similar scheme to analyse the labour market effects of job mismatches among school leavers in Europe. Kivinen and Nurmi (2003) apply a comparable classification to determine the proportion of graduates employed in their ‘own field’ four years after graduation, as well as the proportion of those working in a close field and of those working in an entirely different one. However, this way of operationalising occupational specificity is necessarily a very crude way of assessing the match between educational credentials and labour market outcomes, since both the classification of fields of studies and the classification of matching occupations are very heterogeneous in their single categories. This becomes most apparent for the subject cluster “social science, business, law”, which not only incorporates three broad subject categories, but also groups a strongly profession-specific subject such as law together with the more general subjects of social sciences. The large number of occupations matching this field of study group is not surprising, given the large variety of subjects included. A simple way out of this problem would have been to use a more refined classification of fields of study. However, this study had to make use of the broad subject clusters, not only because the low number of cases in the German data set might have caused problems for fitting multivariate models, but also because the older British cohort (NCDS) data-set only contained field of study information on such a broad basis. Another problem of this way of measuring occupational specificity is the assignment of occupations to fields of study based on the researcher’s own subjective judgement of congruence, which might lead to biased assignments based on the researcher’s nationality and background. Since the classification used in this study is very similar to those used by Wolbers (2003) and Kivinen and Nurmi (2003), researchers with both different study interests and different nationalities, the problem of subjective bias is considered low due to the high degree of correspondence of the single categories. A more objective way of approaching the congruence between field of study and occupation and, furthermore, of incorporating national specificities of this relationship would have been to calculate an occupational dispersion index for each field of study held by graduates. This method has been applied by Dekker et al. (2002), who were interested in the opposite direction, namely the educational dispersion of single occupations. They computed the Gini-Hirschman dispersion index of the educational background of workers in the occupations concerned by means of the Dutch labour force survey. An application of this approach to the estimation of dispersion of occupations in single fields of study can be expressed as OSS = (1-∑ (NWSO/NWS)2) * (O/(O-1)) with OSS = dispersion index of occupations in field of study s; NWSO = number of graduates in occupation o with field of study s; NWS = number of graduates with a higher education degree in field of study s; and O = number of occupations. This approach would most probably yield better estimates for assessing the match between fields of study and occupations in different national contexts. However, this study could not make use of such a measurement, again due to the broad subject clusters adopted, since the dispersion of occupations in each field of study category necessarily has to be rather high. However, in data sets where more refined information on fields of study is available, the 95 physicists, chemists, mathematicians, and statisticians (codes 211 and 212); the category of Health/Welfare includes health professionals (code 222), but also nursing and midwifery professionals (code 223), since both are taught as subjects at British universities. The following table summarises the ISCO 88 occupations that match a particular subject. To determine the impact of occupation-specific training on occupational mobility taking place after initial employment, the study applies a competingrisk model that disaggregates job-shift rates within and between occupations. In particular, much of the analysis rests on a broad reclassification of the original three-digit ISCO-88 codes into eight broad occupational groups. This classification distinguishes between (1) management occupations, (2) professional occupations, (3) technicians and associate professional occupations, (4) clerks, (5) service and market sales personnel, (6) craft and related trades workers, (7) plant and machine operators, and (8) elementary occupations including skilled agricultural workers. It mostly represents the broad ISCO-88 occupational groups, which is a compromise between established practices, feasible data harmonisation, and a level of disaggregation that still permits meaningful statistical modelling, in particular with the smaller German sample. Any occupational mobility processes occurring below this particular level of aggregation could therefore not be taken into account. Labour Market Segmentation To operationalise the degree of labour market segmentation at the institutional level, professional and public labour market segments were analysed in relation to their historical development, entry level requirements, career mobility patterns, and the quantitative share of people employed in these segments for both countries. Special attention was paid to the countryspecific legal definition, the importance of higher education credentials as entry certificates as well as the degree to which both segments are sheltered from market mechanisms. To analyse the importance of OLM and ILM segments for graduate career development at the individual level, the labour market of higher education graduates was divided into four exclusive segments along the axes of public/private and professional/non-professional, as suggested in the theory chapter. The operationalisation of the professional labour market segment was based on the ISCO 3-digit category 200 “Professionals”, consisting of the sub-categories 210 “Physical, mathematical and engineering science professionals”, 220 “Life science and health professionals”, 230 “Teaching professionals”, and 240 “Other professionals”, such as business professionals (241) or legal professionals (242). calculation of an occupational dispersion index for each subject is likely to produce a better measurement of the occupational specificity of higher education systems. 96 Measurement of public sector employment could not be based on a comparable indicator, but had to rely on the variables provided in both data sets. This necessarily results in a higher chance of bias in estimates based on country-specific differences in the definition and composition of public sector employment, since the coverage of public sector statistics is usually determined by national legislation and reflects the historical development and the evolution of the status of public institutions (Hammouya 1999; Rothenbacher 2001/2002). However, in absence of any standardised classification tool, no other possibility was available. In addition, the incompleteness of the data did not allow for the inclusion of some standard measures commonly used in labour market segmentation analysis in the models. For example, it would have been worthwhile to take into account firm size as an indicator of the existence of strong firm-internal labour markets in any of the four segments. Likewise, it was only possible to model segment mobility as between-firm mobility, while within-firm mobility could not be considered. While the SOEP would have included all necessary variables, they are not available in the British data sets. Labour Market Regulation For the operationalisation of labour market regulation at the macro level, unemployment and employment protection legislation were analysed for both countries, with a special focus on reforms that took place during the 1980s and 1990s. In addition, the development of unemployment and employment rates throughout the last 20 years were used quantitatively to assess the effects of the strictness of employment protection legislation. To operationalise the influence of labour market regulation on graduate career mobility at the micro level, several measurements were applied. In order to examine the stability of career development during the first five years after graduation, the mean duration of various spell types as well as the number of full-time and part-time jobs were calculated. The incidence of unemployment was analysed by estimating Kaplan-Meyer survivor function for the transition to unemployment, while the effects of unemployment for further career mobility were approached by comparing the occupational group of a job held before and after spells of unemployment. For this analysis, the eight occupational categories applied were the same as for the estimations on occupational mobility, namely management occupations, professional occupations, technicians and associate professional occupations, clerks, service and market sales personnel, craft and related trades workers, plant and machine operators, and elementary occupations including skilled agricultural workers. Upward and downward status mobility was operationalised by comparing the EGP classes of jobs held before and after job shifts and phases of unemployment. It would have been worthwhile also 97 to include other measures of labour market regulation into the models, such as type of employment contract (fixed-term vs. permanent) or selfemployment. Unfortunately, the former could not be taken into account due to data limitations in the cohort studies, while the latter was dismissed because the number of self-employed individuals in Germany was too low. Control Variables at the Macro and the Micro Levels In order to account for changes in the macroeconomic environment that took place during the observed period, four main indicators were included in most of the models. Variations of labour market demand are operationalised roughly through annual unemployment rates for each country as well as annual changes of the gross domestic product. Both variables were obtained through the OECD Employment Statistics online data base and merged as time-varying covariates to the individual data sets for each year. In order to account for the degree of higher education expansion and thus competition at career entry, the size of birth cohorts and the annual age participation rate were included in the models. The first one was matched with the individual based on the year of birth for each respondent, while the latter was matched based on of the year of graduation. Even though it would have been worthwhile to include further indicators, such as the size of the public and professional sector or the strictness of labour market regulation, this approach could unfortunately not be realised since statistical information on country-level development was not available for either country for a period of more than 20 years. In case when including all of these variables was not feasible due to problems with fitting models, graduates were divided into two graduation cohorts, the first consisting of graduates receiving their degree prior to 1990 and the second of graduates leaving the university in 1990 and later. This division controls for changes in both countries contrasting the 1980s with the 1990s, but also allows for considering changes stemming from German reunification. As will be shown by the institutional analysis, changes that are captured by this differentiation are variations in the size of the student and graduate population (as a result of educational expansion in both countries), changes in size of the public sector employment in both countries, changes in the regulation of the professional labour market (most pronounced in Britain), and changes of labour market regulation in both countries (again most pronounced in Britain). Of course, this way of assessing continuity and change is less accurate than including relevant macro variables into the respective models, since the different forms of social change cannot be disentangled any longer. Control variables at the micro level consist of general socio-structural indicators such as gender, dependent children under the age of six, the 98 parental socio-economic status/education, and nationality/ethnicity. Due to data restrictions and missing values, the parental socio-economic background as well as ethnicity had to be coded differently for each country. While, in Germany, parental socio-economic status is measured by parental educational attainment, namely whether the father possesses a higher degree at the time the respondent is 15 years of age, in Britain, parental background measures whether the father is occupied in EGP class I or II at the time the respondent is 16 years of age. Ethnicity in Germany is coded as German vs. Non-Ger- man, while ethnicity in Britain is coded as white vs. non-white. Despite knowing that these variables measure different socio-structural aspects in both countries and are thus of limited use for direct comparison, they have been included in order to control for within-country variations. In addition, further indicators related to human capital and indicators of labour market experience were included in the models. Some of them are more general in nature, e.g. work experience, vocational education in addition to higher education, or the duration of unemployment spells. Others are related to higher education, but are not specifically captured by one of the institutional concepts. For example, a variable measuring graduation in East Germany controls for differences between West and East Germany after reunification. The age of graduation was also controlled for since it differs strongly between the two countries. It was included in the form of age intervals (below 24 years, 24 to 29 years, 30 years and over) to allow for curvilinear age effects. 99 5. Higher Education and Labour Market Institutions in Germany and Britain This chapter examines higher education and labour market institutions in Germany and Britain to explore how they might shape graduate careers. In order to analyse higher education stratification, an account is given of the legal basis and structural set-up of the German and British higher education systems, and explicates the educational pathways that are necessary preconditions for entry into higher education. The chapter then goes on to explain the importance of occupation-specific contents of higher education programmes for labour market entry by analysing training for occupations in general as well as specifically for professionals, engineers and managers in both countries. Also, it sketches in rather broad terms the different principles underpinning higher education in both countries and elaborates how these different principles have created country-specific cognitive frameworks for the relative importance given to occupation-specific and general higher education. The different segments of the graduate labour market are taken into consideration by looking at the way professions and the public sector are organised in both countries, with a particular focus on their social closure from external labour markets. In order to account for different modes of labour market regulation, the institutional analysis proceeds by examining how national systems of labour market protection are organised. A short historical overview on the development of institutional features is given at the beginning of each section. In addition, each institutional analysis is accompanied by a quantitative assessment of relevant indicators. Instead of giving a mere descriptive account, the aim of this comparative analysis is to link the set-up of higher education and labour market institutions back to the discussed theoretical concepts in order to examine how national institutions might structure higher education to work transitions. This allows for making more explicit hypotheses about transition patterns for both countries, which will be tested afterwards by the quantitative analysis. But for now, the focus is on the more qualitative study of higher education and graduate labour markets in Germany and Britain. 100 5.1 Stratification in Higher Education: Vertical and Horizontal Institutional Differentiation Stratification of higher education systems depends on their vertical and horizontal institutional differentiation in combination with the stratification of secondary education and the degree of a country’s higher education expansion. The main assumptions connected with this concept are that the lower the percentage of a birth cohort that acquires higher education and the more vertically differentiated tracks exist, the more strongly are graduates channelled into particular sections of a stratified occupational structure. Different types of higher education institutions, degree levels, as well as fields of study have been identified as indicators of institutional differentiation within a higher education system (Teichler 2002a). The analysis of these institutional structures in Germany and Britain will demonstrate whether they are differentiated vertically or horizontally and, based on this differentiation, whether they are likely to stratify individual labour market outcomes. But not only does the structural set-up of higher education itself matter. The structure of secondary schooling and resulting pathways into higher education are important aspects of institutional stratification as well, since these strongly influence higher education participation rates. Before analysing the structural set-up of both levels of education in Germany and Britain in detail, this chapter gives an overview of the development of higher education expansion. 5.1.1 The Road to Mass Higher Education The overall goal of higher education policies in Germany and Britain nowadays is to increase participation in higher education and thereby foster higher education expansion. The British Labour Party explicitly aimed at achieving an enrolment rate of 50 percent of those aged 18–30 by 2010 (DfES 2003). Even though it may not have such explicit figures in mind, the German federal government also stressed the importance of increasing the number of students in higher education (BMBF 2007). And indeed, both countries have experienced a massive expansion of higher education during the past decades. Initially, higher education expansion in Germany and Britain was not merely a political process, but was mainly driven from below. Ongoing processes of democratisation, which enabled ever larger numbers of citizens to participate in society and to profit from the expansion of welfare programmes, lead to an increasing demand in education participation among all social groups (Klemm 2000; Müller 1998). The German saying “Send your child to better schools for a longer period of time” (“Schick dein Kind 101 länger auf bessere Schulen”) is a good example of the rising value of education (Klemm 2000: 150, own translation). While rising participation in education naturally started at lower educational levels, such as primary and secondary schooling, soon more and more children acquired upper secondary credentials, the necessary prerequisite for entry into higher education. As a consequence, demand for attending universities afterwards rose considerably. Given the relatively low number of university institutions in both countries at the beginning of the last century, education policy had to respond to the expansion of education. For the decades following the Second World War, German and British governments increasingly responded to the enormous education demands by increasing public spending as well as introducing new institutional structures. First political attempts to accommodate the increasing number of students started shortly after the Second World War in Germany and Britain, even though the overall goals of higher education policy differed as a result of the respective post-war experiences (see Eurydice 2000). In Germany, the primary focus was on reconstructing and re-establishing German higher education after the end of the National Socialist era, while participation rates only played a minor role. The organisational and structural set-up of German higher education institutions was reinstated and in many respects replicated from the situation existing prior to 1933. Initially, the German Federal Government exerted no influence whatsoever on this development, since according to the principle of cultural sovereignty (Kulturhoheit) as reinstated in the Constitution the rebuilding of the higher education system was a matter for the Länder.17 Since 1948, Länder policies on higher education were 17 The principle of federalism (Föderalismus) in Germany may be understood against the background of Germany's constitutional and state tradition. One of the fundamental elements of the Basic Law (Grundgesetz), besides the principles of democracy and the rule of law, is the principle of federalism (Article 20, Paragraph 1). Its major characteristic is that both the Federation and its constituent states, known as Länder, have the status of a state. One core element of this status is, according to the constitutional order laid down in the Basic Law, so-called cultural sovereignty (Kulturhoheit), i.e. the predominant responsibility of the Länder for education, science and culture. This element lies at the heart of their sovereignty and means in principle that each Land bears responsibility for its educational and cultural policy. On the other hand, the constituent states of the federal state bear joint responsibility for the entire state. This overall responsibility both entitles and obliges them to cooperate with one another and to work together with the Federal Government. Federalism and cultural sovereignty have a long constitutional tradition in Germany. Various models of state organisation developed within the framework of the federal order include the Holy Roman Empire of the German Nation (to 1806), the German Confederation (1815-1866), the German Empire (1871-1918) and the Weimar Republic (1919-1933). The founding fathers of the constitutional order established by the Basic Law created a federalist order in the newly-founded Federal Republic of Germany in 1949 not only in order to carry on a constitutional tradition, but also in order to make a conscious break with the National Socialist centralist state (1933-1945). In doing so, they particularly 102 coordinated by the Standing Conference of the Ministers of Education and Cultural Affairs (Kultusministerkonferenz KMK). This strong federal component in German higher education governance changed with the beginning of higher education expansion. University enrolment first began to rise appreciably in the mid 1950s, a trend that continued into the 1960s against the background of public concern about an imminent shortage of graduates and the call for the reduction of social and regional inequalities. The political response was to expand existing universities and establish new ones in structurally weak regions. A key impetus was provided by the Science Council (Wissenschaftsrat), set up in 1957, in which the Federal Government and the Länder worked together for the first time and which served to institutionalise the cooperation of academic and governmental actors in higher education. In Britain, the first phase of higher education expansion started before the end of World War II, when more and more families from all class backgrounds demanded better educational opportunities. In response and as a first step towards higher education expansion, the Education Act (1944) proposed to educate and mobilise the working class by abolishing tuition fees charged in the British secondary school system. As a result, a far higher percentage of pupils attended upper secondary schools, which led to a vast increase in the number of qualified young people entering higher education. Only two years later, the Barlow Report (1946) worked in the same direction by recommending a doubling of university student numbers, especially in science subjects, to meet the country’s need for scientific manpower. As a consequence, both government funding and student numbers were greatly increased in the immediate post-war period in Britain. During the 1960s, both countries experienced a continuously growing demand for higher education participation. As a consequence, education policy more and more focussed on the expansion of higher education, which subsequently lead to the establishment of a binary structure of higher education institutions. In Germany, the post-war consolidation period was followed by a phase of active higher education development as a response to the increasing number of students. From around 1960 onwards, West German higher education institutions were under considerable pressure to change and modernise, which, in turn, initiated a phase of fundamental reform in the higher education sector. Public spending for the higher education system rose significantly. Between 1965 and 1980, the structure of the higher education system changed considerably, partly due to the introduction of new types of higher education institutions and partly due to the merging of existing ones. The constitutional provision for these reforms was fleshed out by the 1969 Higher Education Institutions Construction Act returned the education system into the hands of the Länder. This federal order was also retained after the establishment of German unity in 1990 (Eurydice 2006). 103 (Hochschulbauförderungsgesetz), a law to promote the joint task of expanding existing institutions of higher education and building new ones in order to cope with rising enrolment rates (BMBF 1969). One of the most prominent reforms was the establishment of Fachhochschulen (technical colleges), which were introduced in 1970/71. The characteristic feature of course design as well as teaching and study organisation at this new type of institution was and continues to be a particular emphasis on the practical application of knowledge and an orientation towards labour market requirements. With the establishment of Fachhochschulen, German higher education moved towards a binary structure within its higher education system, consisting of universities and technical colleges (Eurydice 2000), a structure which remains in place until today.18 In Britain, the continuously rising number of higher education students created a pressure for change in the provision of higher education in the early 1960s, too. The subsequently established Committee on Higher Education made a comprehensive assessment of the present and future of British higher education. The resulting Robbins Report (1963) laid down the basic principles of higher education, which guided university development in subsequent years. Its main goal was to provide higher education courses to all those who wished to do so, as long as they were qualified in terms of their abilities and credentials (Robbins 1963). Moreover, it recommended further expansion, a broadening of both the regional spread and of the scope and diversity of university education. This was accepted by the Government, and additional funds were made available to the University Grants Committee (UGC) to meet the new targets. In the following years, new universities were created, and many technical colleges were upgraded to receive university status. However, just as in Germany, it was recognised that there was a growing need for vocational, professional, and industry-based training, which could not be fully met by the universities. As a consequence, the White Paper “A Plan for Polytechnics and Other Colleges”, published in 1966 (DES 1966), aimed at establishing non-university higher education by introducing a new type of institution, the so called polytechnics. The subsequent foundation of 31 polytechnics in England, Wales and Northern 18 In addition to this primary institutional differentiation, the German university system consists of a range of institutions that also enjoy university status. Besides the traditional universities, there are universities of technology (Technische Hochschulen or Technische Universitäten) that specialise in the natural and engineering sciences, or schools that only offer a limited range of degree courses, for example theological colleges and Pädagogische Hochschulen (colleges of education). The 29 Fachhochschulen for public administration (Verwaltungsfachhochschulen) play a special role, which train civil servants for careers in the so-called higher level of the civil service. They are maintained by the Federation or by a Land, and their students have revocable civil servant status. However, the main differentiation in German higher education is between university-type institutions and technical colleges, as will be demonstrated in the following. 110 improve higher education management. This has involved, for example, a review of Regelstudienzeiten (standard periods of study) and examination requirements at both types of institutions in conjunction with improvements in teaching. One priority still is to expand Fachhochschulen and to make them even more attractive, for example by consolidating applied research work and technology transfer (Eurydice 2000), which again indicates a more horizontal differentiation between the two types of institutions. Furthermore, institutions of higher education are to be made more efficient by granting them further autonomy, allowing them to build an individual profile in a particular area and encouraging more competition. With the amendments of the Framework Act for Higher Education of 1998 and 2002, the Länder enjoy greater scope regarding their own decisions. However, substantial reforms of the binary institutional structure of German higher education have not been on the agenda of higher education reforms. Even the latest revisions of the German higher education system in the wake of the Bologna Process leave the horizontal differentiation between universities and Fachhochschulen largely untouched. In contrast to the stability of German higher education institutions, British higher education has been marked by strong reforms during the same period. From the 1970s until the early 1990s, British higher education was provided by three main types of institutions21: universities, polytechnics, and non-university institutions of higher education. British universities are autonomous institutions, empowered by a Royal Charter or an Act of Parliament. This autonomy from state administration is particularly relevant for course design, since universities have their own degree-awarding powers and determine which degrees and other qualifications they offer as well as the conditions they apply. As a consequence of this organisational independence, British universities are diverse, be it in terms of size, mission, subject mix, and of course history. Generally, universities offer research opportunities as well as a wide range of courses at the undergraduate and postgraduate levels, catering more to labour market requirements, although the balance between these activities varies between institutions. Sometimes universities also offer specific professional qualifications and certain qualifications below degree level. The second type of higher education institution in Britain was the polytechnic, which existed up until 1992. They were originally set up by charitable endowment to enable men and women of lower socioeconomic 21 An analysis of the institutional framework of Britain sometimes has to differentiate between England, Wales and Northern Ireland on the one hand and Scotland on the other, since the latter has followed a somewhat different pathway due to the existence of a distinct education system. To acknowledge this fact does not mean, however, that there are completely different types of higher education in these regions in Britain, but rather that certain aspects vary in their specific organisation, just as they might vary between the German Länder. The most important variations will be stated separately. 111 status to advance their general knowledge and industrial skills on a part- or full-time basis. Thus, similar to their German counterparts, the specific focus of teaching and course design at polytechnics was on applicable knowledge. However, unlike German Fachhochschulen, they could offer the full range of academic qualifications up to doctorates and in general provided some programmes in traditional academic fields. The third type of British higher education institution are higher education colleges and university colleges – so called non-university higher education institutions, which are also characterised by an emphasis on vocational and professional courses. In contrast to university institutions, only a small number of university colleges and higher education colleges have the power to award their own degrees and qualifications. The majority of qualifications offered by non-university institutions are validated by external bodies such as universities or national accrediting bodies. If polytechnics had the right to award degrees, these rights were normally restricted to first degrees and taught (not research) master’s degrees. The differentiation between British universities and polytechnics points to the opposite direction compared to Germany. On the one hand, both types of higher education institution theoretically offered the same type of degrees. However, as regards institutional governance polytechnics did not have the same autonomous status as universities, particularly in relation to degree awarding powers, but were financed and regulated by local education authorities (LEAs). This legal differentiation between autonomous universities and externally controlled polytechnics created a higher education system which differed strongly in quality and quantity, for example as regards the degrees offered, the emphasis placed on research, financial resources, and the volume of public funding. Finally, the universities’ reputation has traditionally been much higher than that of polytechnics (Chevalier and Conlon 2003). As a result, the make-up of the student body was remarkably different, since polytechnics showed a much higher enrolment of students with lower socio-economic backgrounds or from ethnic minorities, and of mature students (Booth 1999). These differences in institutional hierarchy and reputation can be taken as indicators of a vertically differentiated higher education system, which should lead to strong status differences in labour market outcomes of higher education graduates from universities and polytechnics. British higher education policy, however, aimed at putting an end to this hierarchical binary structure as a response to mass higher education (Booth 1999). First changes in this regard were introduced by the Education Reform Act (1988), which removed polytechnics from the control of local authorities and provided them with more autonomous funding via the Polytechnics and Colleges Funding Council (PCFC), a Non-Departmental Public Body. Even more substantial reforms have been proposed by another White Paper in May 112 1991 named “Higher Education: A New Framework” as a response to the increasing number of students in higher education (DES 1991). Its main recommendation concerned the abolition of the “binary system” between universities and polytechnics, leading to the establishment of a unitary system in British higher education. The proposed reforms formed the basis of the Further and Higher Education Act (1992). With this act, polytechnics were given university status if they fulfilled certain criteria such as a specific range of courses offered and a particular distribution of students across curricular areas. When polytechnics met the requirements, they were allowed to include the word ‘university’ in their title and became legally independent, self-governing institutions, with the power to award their own degrees just like universities. By the year 2000, many of the former polytechnics had been granted university status. Today, 88 universities exist in England, Scotland, Wales and Northern Ireland, while there are only 62 non-university institutions (Eurydice 2006) However, regarding the degree of institutional differentiation following from this reform, the binary structure still seems to be in place in the British system (Booth 1999; Eurydice 2000). Despite the upgrading of polytechnics and the creation of one unified sector for British higher education, some of these organisational differences between the ‘old’, or ‘pre-1992’ universities and the ‘new’, ‘modern’ or ‘post-1992’ universities22 remain. Differences continue to exist, particularly with regard to mission and subject mix. Old 22 Within the “old” universities, there are two further lines along which universities are separated. The first one is related to the age of university institutions, which distinguishes between the traditional universities, most of which were established in the Middle Ages, and university institutions founded during the first wave of higher education expansion in the 1960s, which are often referred to as “new” universities. Due to this differentiation within the old universities, another term had to be found for former polytechnics. Therefore, the newest of the new universities, which attained university status only in 1992 by the Higher and Further Education Act, are called “modern” universities in order to differentiate them from the “new” universities. As a result, today three types of universities can be found in Britain today: traditional, new, and modern universities, which all differ in the criteria quoted above. In addition, a more informal divide exists among the old university institutions, one that is related to institutional reputation and research strength. A self selected informal coalition of 19 research intensive institutions formed, encompassing mainly the oldest and most prestigious higher education institutions, but also some of the new universities. This group of institutions is referred as the Russell Group. The members of the Russell Group universities are University of Birmingham, University of Bristol, University of Cambridge, Cardiff University, University of Edinburgh, University of Glasgow, University of Leeds, University of Liverpool, University of Manchester, University of Newcastle upon Tyne, University of Nottingham, University of Oxford, University of Sheffield, University of Southampton, University of Warwick, Imperial College, King's College London, London School of Economics and University College London (Chevalier and Conlon 2003). However, since the data sets used for Britain do not contain any information on institutional differentiation among the old universities, higher education stratification will be restricted to a differentiation between pre-1992 universities on the one hand and polytechnics and non-university institutions on the other. 113 universities still tend to provide more academic courses rather than professional training. New universities, in contrast, generally place greater emphasis on the practical application of knowledge and consequently offer a wider range of courses leading to the professional qualifications recognised by professional institutions. In addition, the amount and the resources of funding differ strongly between modern and old university institutions. Old universities generally obtain a much higher share of state allocated funds. For example, the former polytechnics received only 7 percent of resources allocated through the Research Assessment Exercise, which is one of the major funding sources for British universities today (Hillyard et al. 2003: 10) These distinctions, now working more informally, still reproduce the pre- 1992 institutional hierarchies in reputation and standing. Overall, this hierarchy of institutions continues to create a strong vertical differentiation among both types of British higher education institutions. This strongly stratified British system has already been shown to have an impact on graduate earnings (Chevalier and Conlon 2003). In addition, labour market rewards after graduation should be strongly stratified between traditional universities on the one hand and former polytechnics and other nonuniversity institutions on the other.23 Institutional Differentiation by Type of Degree and Field of Study The types of degrees awarded in German and British higher education vary even more between the two countries than the types of higher education institutions. But again, institutional differentiation follows a similar pattern as above, with German degrees exhibiting more horizontal demarcations, and British degrees clearly being vertically structured. In addition, both higher education systems are differentiated horizontally, mainly along faculties or 23 Somewhat different to England and Wales, four different models of governance in the higher education sector existed in Scotland until 1992, each reflecting different founding origins. But despite these differences, the main institutional differentiation between traditional universities and polytechnics/non-university institutions can be found there as well. The four ‘ancient’ universities (University of Aberdeen, University of Edinburgh, University of Glasgow, St. Andrew’s University) were founded as legally autonomous governing bodies by the Scottish Universities Acts in 1858 and 1889 and are best comparable to the traditional English universities. Another four Scottish universities (Dundee University, University of Stirling, Strathclyde University, Herriot Watt Edinburgh) were established by Charter during the 1960s and most closely resemble the new universities in England. Both types of institutions enjoyed university status and thus institutional autonomy from their beginnings. Designated institutions, on the other hand, were similar to English polytechnics. They are governed in accordance with Orders of Council made under Section 45 of the Further and Higher Education (Scotland) Act 1992 and only then received university status. Very much like English non-university institutions, the Scottish colleges of education are governed in accordance with the Colleges of Education (Scotland) Regulations 1987. 114 departments of different fields of study. These cross-national variations in institutional differentiation by degree types and fields of study will be described in more detail below. In addition, the institutional analysis will be linked to the Casmin scheme, which is supposed to distinguish educational levels according to their decisive signals for utilisation on the labour market and their functional equivalence of its education categories across countries (Brauns et al. 1997b). To begin with, undergraduate degrees did not exist in Germany prior to the sweeping Bologna reforms, and all degrees were conferred at the postgraduate level. The lack of a hierarchy among different degree levels signals a horizontally differentiated system. Moreover, all degrees conferred at German higher education institutions used to be organised and coordinated by the same regulatory framework. To ensure that the various institutions of higher education throughout the country provide a comparable standard of scientific and academic training and degrees, the KMK and the Association of Universities and other Higher Education Institutions (Hochschulrektorenkonferenz HRK) set up a Joint Commission for the Coordination of Study and Examination Regulations, which has drawn up a framework regulation, covering all subjects and degrees available in the German higher education system. The framework regulations contain the quantitative reference data for courses of study, in particular the Regelstudienzeit (standard period of study), the amount of hours of teaching on required and elective subjects, the number of certificates required for admission to examinations (Leistungsnachweise), examination details and the length of time allowed to complete the final dissertation.24 The only differentiation existing among German degrees in the period studied in this analysis is a horizontal one, associated with different fields of study and labour market segments. At university-type institutions, the most 24 These frameworks regulated all matters related to degree and field of study organisation until the end of 2002 when the changes proposed by the Bologna Declaration were introduced. Following from a KMK resolution in March 2002, a system of accreditation will replace the previous system of coordination of study and examination regulations in the future. For the accreditation of Bachelor’s and Master’s study courses, the KMK has set up an accreditation council (Akkreditierungsrat) acting on behalf of all Länder. In the long run, the system of accreditation will be extended to all courses of study. Furthermore, the KMK issued structural guidelines for Bachelor’s and Master’s degree courses that have already been incorporated into the Framework Act for Higher Education in October 2003. These rules, which are facultative for all Länder, regulate, amongst other issues, the structure and length of the study. The structural guidelines stipulate that Bachelor’s study courses, as study courses which lead to a first degree qualifying for entry into a profession, must provide the academic foundation, methodological skills and qualifications related to the professional field. Furthermore, the structural guidelines distinguish between more research-oriented Master’s study courses and more practice-oriented ones. Bachelor’s and Master’s study courses are provided with a credit point system which is based upon the European Credit Transfer System (ECTS). 115 important distinction is made between academic degrees (Diplom and Magister) and state examinations (Staatsexamen). The general Regelstudienzeit for most of these degrees is eight to ten semesters (with the exception of medicine). Academic and state examinations do not build upon each other; rather, professional qualifications are conferred on the basis of all of them. They provide for a horizontal differentiation because they vary in the degree of specialisation on specific subjects and the number of subjects studied. State examinations have to be taken in degree courses that prepare students for professions of particular importance to the public interest, such as medicine, pharmaceutics, law, or the teaching profession. Academic degrees, in contrast, are not considered to prepare for such a specific occupational field or profession, but are more generally related to specific research disciplines, such as mathematics or sociology. Thus, the difference between the two types of degrees essentially concerns their content and future occupational domain. Their curricular structure and type of examination are quite similar, with the main exception that in the case of Staatsexamen, where representatives of state examination bodies act as examiners in addition to university professors. Within the academic degrees, a further differentiation exists. A degree is awarded on the basis of either the Diplomprüfung, which leads to the awarding of the Diplomgrad, or the Magisterprüfung leading to the awarding of the Magistergrad. The main difference between the two is again related to the fields of study, since Diplom programmes concentrate on a single subject, while those that lead to a Magister degree allow for a combination of several subjects, usually one major subject and two minor subjects, or two equally weighted major subjects. Both types of academic degrees are offered in a large range of subjects including the humanities, economics, social sciences, sciences, or engineering. However, the Diplom is more often awarded for subjects with a stronger focus on applicability such as engineering, while Magister degrees are generally conferred for subjects more general in nature, such as humanities or arts. Even though all of these degrees are already conferred at a postgraduate level, further postgraduate degrees can be obtained in Germany following successful completion of a first degree. For instance, particularly wellqualified students, who have achieved a certain level of academic performance, may choose to complete a doctorate (Promotion). A doctorate is conferred on the strength of a doctoral thesis, which must be based on independent research, and an oral examination (Rigorosum) or alternatively the oral defence of the thesis (Disputation). In addition, postgraduate degrees of further, supplementary or follow-up courses (postgraduale Studiengänge) can be taken. Unlike further education taking place at the work place, 116 postgraduate degrees are usually pursued immediately after or even parallel to first degree courses.25 The main distinction between academic and state degrees can be taken as an indicator for a horizontal institutional differentiation in the German university sector, since they are rather equal in terms of study organisation, duration of study, and degree level. Also, the various degrees conferred at German universities are generally classified as Casmin 3b level degrees – higher tertiary education. In this regard, the Casmin classification does not distinguish any further between state and academic examinations or between first and second stage postgraduate degrees. This can, on the one hand, be taken as another indicator that German university degrees differ more horizontally than vertically, but on the other that Casmin is not precise enough to capture the specificity of German degrees. As a consequence, labour market outcomes should be weakly stratified between academic and state degree holders. Fields of study are generally considered to differentiate the student population horizontally and should therefore support a low stratification of labour market returns. In contrast to universities, German Fachhochschulen only award one kind of degree, the Diplomgrad, upon completion of a course, which normally lasts for four years. But just like university degrees, Fachhochschule degrees are only conferred at the postgraduate level, without any intermediary examination which would qualify for the labour market. Since the characteristic feature of this type of institution is an orientation towards the requirements of the labour markets, all fields of study taught at Fachhochschulen have to show a particular emphasis on applicability and practical knowledge, while aspects of basic research and theory advancement are of less importance. The subject canon in parts differs from the one taught at university institutions, comprising for example social work, social services, public administration, administration of justice, design, nursing, or public health management. Upon completion of a first Fachhochschule degree, further studies, supplementary and follow-up courses can be taken as well, which last two to four semesters and culminate in the award of a second Diplom degree or proof of academic achievement (certificate). It is not possible to obtain a doctoral degree from Fachhochschulen, given that only universities and equivalent institutions of higher education are entitled to award doctorates. 25 Unfortunately, information on degrees in the SOEP does not include such postgraduate degrees and therefore the main differentiation made in the individual analysis is between academic and state examinations. Bachelor's and Master's degrees, which are now available at German higher education institutions, are not included in the examinations of individual career trajectories either, since the implementation of the Bologna Process is still ongoing and since this study focuses on labour market entries prior to its beginnings. 117 The inclusion of Fachhochschule degrees in a system of institutional differentiation poses some difficulties. On the one hand, the organisation of studies and the associated fields of study differ strongly between both types of higher education institutions. Based on these differences, degrees awarded by Fachhochschulen have often been considered inferior to university degrees. In the Casmin scheme, they are classified as lower tertiary degrees in comparison to higher tertiary degrees awarded at universities (Brauns et al. 1997b), which signals a vertical differentiation between university and nonuniversity degrees in Germany. However, the picture changes if the degree level and consecutiveness of degrees are taken into account. At the degree level, Fachhochschule degrees are rather similar to university degrees, since all are conferred as postgraduate degrees. In addition, Fachhochschule and university degrees do not constitute a two-tiered structure of degrees, with one degree building upon the other, as it is the case in Britain. They rather constitute different degree options at the same level, between which students have to choose at the beginning of their higher education studies. Studying for both types of degrees simultaneously or consecutively is uncommon. Based on these latter characteristics, it is also possible to classify the German degree system as being horizontally differentiated. As a consequence, labour market outcomes should not differ too strongly between university and Fachhochschule graduates. Rather than the mere differentiation by different types of degrees, field of study might matter in this regard, since all university and Fachhochschule degrees are also differentiated by the subjects associated with them. In Britain, degrees can be divided into undergraduate and postgraduate level degrees at all types of higher education institutions, constituting a clearly vertically differentiated degree system. At the undergraduate level, two different kinds of degrees exist, so called intermediate level degrees and Bachelor degrees. Higher Education Diplomas and Certificates are intermediate level degrees and the shortest types of degrees available, lasting for a maximum of one to two years. However, they are recognized as higher education qualifications in their own right. Certificates of Higher Education normally take one year of full-time study and the holder is expected to have a sound knowledge of the basic concepts of a subject, and will have learned how to take different approaches to solving problems. Higher Education Diplomas usually take two years to complete by full-time study. Holders of these qualifications are expected to have developed a sound understanding of the principles in their field of study, and will have learned to apply those principles more widely than Certificate holders. Both types of degrees are often delivered through partnerships of further and higher education institutions, and are intended to help education providers to address the shortage of intermediate-level skills, to widen participation in higher education, and stimulate lifelong learning. They are usually offered for more 118 vocationally oriented subjects with a more vocational orientation and in employment-related subject areas such as internet computing, learning support, or hospitality, leisure and tourism. According to the Casmin scheme, these intermediate level degrees are classified as lower tertiary degrees (3a), just as German Fachhochschule degrees (Brauns and Steinmann 1997). The second undergraduate stage in Britain lasts for three or four years and leads to a Bachelor of Arts or Bachelor of Science Degree (B.A./B.Sc.), which can be taken as a first degree or build upon an intermediate degree.26 Graduates with a Bachelor degree are supposed have developed an understanding of a complex body of knowledge, some of it at the current boundaries of an academic discipline. The Bachelor’s Degree is conferred as a Pass Degree for more general studies or an Honours Degree where studies are more specialised. Honours degrees form the largest group of British higher education qualifications. Typically, courses last for three or four years (if taken full-time) and lead to a bachelor’s degree with titles such as Bachelor of Arts (B.A.(Hons)) or Bachelor of Science (B.Sc.(Hons)). Longer courses include sandwich courses, which incorporate periods of practical work in organisations outside the university or college, and courses specialising in modern foreign languages, which normally include a year abroad in the target language country. Shorter courses include accelerated two-year degrees, which normally require students to study during the normal vacation periods. The majority of Bachelor students are expected to have identified their field of study when they apply for admission. Most programmes specialise in a specific subject area, similar to the German subject canon, where that subject is set in its broader context. However, it is also possible to undertake a combined studies programme involving two or possibly three specialisations. Typically, there is a relatively fixed menu of course modules which are intended to cover the core knowledge of the subject, with an element of choice which increases in the later years of the course of study. Some programmes offer a wider range of subject areas providing a good advanced general education. In the Casmin scheme, Bachelor degrees are typically classified as higher tertiary degrees (3b). However, the more refined version of this scheme allows for taking into account their undergraduate status, categorising them as lower level 3b degrees (3b_low) (Brauns and Steinmann 1997). 26 The main difference between English and Scottish Bachelor degrees is related to their length, since Scottish honours degree courses generally last for four years in comparison to three years of study in England. This is mainly related to the younger leaving age of Scottish students from upper secondary education (see section 5.1.3. on ‘Pathways into higher education’, Footnote 28). Another Scottish specificity is related to the fact that the first degree in Arts in the four ‘ancient' universities and Dundee University is the Master of Arts. MA degrees in all other faculties as well as in the other universities are conferred as post-graduate qualifications. 119 At the postgraduate level, a Master’s Degree is conferred after one or two years of study in a specialised field following the Bachelor’s Degree.27 Much of the study undertaken at the Master’s level will have been at or informed by the forefront of an academic or professional discipline. Master’s degrees are awarded after completion of taught courses or programmes of research, or a combination of both. Common degrees obtained at this level include Master of Arts (MA), Master of Science (MSc), Master of Business Administration (MBA), Master of Education (MEd), Master of Social Work (MSW), Master of Musical Arts (AMusM), or Master of Medical Sciences (MMedSci). Some Master’s Degrees – for example in science and engineering – are awarded after extended undergraduate programmes that last, typically, a year longer than honours degree programmes. Advanced short courses, also situated at this level, often form parts of continuing professional development programmes, leading to postgraduate certificates (PGCE) and postgraduate diplomas. While a lot of Master’s Degrees aim at teaching applicable knowledge for the professional life, the Master of Philosophy and the Doctor of Philosophy can be considered pure research degrees at the postgraduate level. Following an honours degree, the Master of Philosophy (MPhil) Degree is awarded after two years of additional study and the successful presentation of a thesis. After usually three years of further study beyond the Master’s Degree, the candidate may present a thesis for the Doctorate of Philosophy (DPhil or PhD). Both of these titles are awarded for the creation and interpretation of knowledge, which extends the forefront of a discipline, usually through original research. In addition, there are doctoral programmes that have a substantial taught element and only include a minor research component, usually leading to awards which include the name of the discipline in their title (such as EdD for Doctor of Education). Any type of postgraduate qualification awarded in the British higher education system is classified as upper tertiary degree (3a) by the Casmin scheme, often with the extension of higher level (3b_high). Thus, contrary to the differentiation between German university degrees, which cannot be captured by this cross-national classification, all British degrees have separate categories. Taking all different degree levels together, both Casmin and the explications above lead to the conclusion that the British degree system is strongly vertically differentiated into undergraduate and postgraduate degrees, with most of them building upon each other. Fields of study in Britain are less related to specific degree courses than in Germany. 27 At Oxford and Cambridge, the Master of Arts (MA) is conferred automatically after a certain period of time to all holders of Bachelor's Degrees. In other fields, it is awarded under the same conditions as in other universities. Some Master's Degrees are awarded as undergraduate degrees after four to five years of study (for example at the Imperial College, London), if the Bachelor's programme is included in the programme. 126 labour market entry has been more turbulent, marked by a higher prevalence of unemployment and sequences of stop-gap jobs.29 These country-specific differences of occupation-specific training might exist in the field of higher education as well. However, measurement of the concept at this education level requires some adaptation, since no clear-cut occupation-specific entity such as an apprenticeship exists. Arguably, the most important factor for determining the occupational specificity of a country’s higher education system is the analysis of different fields of study, which can directly be examined in relation to the occupation-specific content taught and the applicability of knowledge for the labour market. Generally speaking, this is very much the case in Germany, where the respective field of study strongly determines the possible range of occupations to work in. In Britain, in contrast, enrolment in a certain field of study traditionally has not been perceived to be a clear decision for a certain occupational area (Teichler 1999: 32). Occupational specificity is also indicated by the incorporation of practical training phases in particular degree courses or by the extent to which particular degree types or fields of study are the necessary prerequisites for work in particular occupations. In addition, the general principles underpinning higher education are considered important empirical indicators of occupational specificity. In the following sections, all dimensions of institutional differentiation will be analysed as regards the importance of occupation-specific training. As a starting point, the general objectives associated with higher education in both countries are described, which already indicate how the link between higher education and labour market is perceived in Germany and Britain. 5.2.1 Historical Frames: Professional vs. Liberal Higher Education German and British higher education systems have emerged from very different historical traditions as regards the applicability of knowledge for labour market purposes and the training of occupation-specific skills. The ideas associated with higher education in both countries can be traced back to the very beginnings of university establishment in Europe in the 12th and 13th centuries. At this time, the two first universities in Europe were founded in Bologna and Paris. These new types of education institutions offered certificates and credentials and thus provided the first form of organised higher education. However, the focus of both universities could not have been more diverse. While the University of Bologna specialised in the education of law and, to a lesser extent, in medicine, i.e. professional knowledge, the University of Paris had a focus on general education by specialising in courses of the artes liberales, such as philosophy, theology, 29 For a short description of German and British vocational training systems see Appendix D. 127 logic, or astronomy (see Moraw 2005; Verger 1993 for the following). These different directions of higher education can be explained by different historical circumstances that accompanied the foundation of universities in both cities. The institutionalisation of higher education in Bologna was stimulated by the ongoing emancipation of Italian merchants from the overall dominance of the Catholic Church in educational matters. In order to gain more political power, independent of Christian sovereignty, and to improve the knowledge about successfully governing the flourishing merchant cities of Northern Italy, the focus of university education tended to be on applicable knowledge. The aim was to educate young men from the nobility for leading positions in public administration, and studying law was considered to be the best preparation to this end. In France, the Catholic Church still had a monopoly in educational matters. Higher education at the University of Paris was not aimed at professional training at all, but rather to develop the intellect and the personality of the young noblesse by educating them in the subjects of the trivium (grammar, rhetoric, logic) and to a lower extent of the quadrivium (arithmetic, geometry, astronomy, music) as a preparation for a clerical career (Leff 1993). In cooperation with the Catholic Church, the University of Paris provided this general and liberal education and soon became the intellectual centre of Europe. In this regard, it is important to note that in contrast to the legal education provided by the University of Bologna, the study of artes liberales was provided at an undergraduate level as preparation for the study of theology (Moraw 2005: 32). After the foundation of these first two universities – the question which is the older one remains unsolved and disputed – other regions in Europe started to adopt this principle of formal higher education through universities. In England, the first universities, those of Oxford and Cambridge, were founded soon after Paris and Bologna. Both universities evolved as private bodies during the thirteenth century and followed the French tradition of liberal education specifically by focussing on philosophy and arts to prepare their students for careers in either the Church or, to a much lower extent, in public administration (Moraw 1993). Rather than preparing their students for particular professions, the “civil conversion of the gentleman” was the most prominent idea behind these early forms of higher education in Britain. The predominance of such a liberal education at the undergraduate level and the importance of personal development rather than professional education remains a distinct feature of higher education in Britain today. Professional training remained outside the British university system for a long time, even though a small minority of institutions, such as the Inns of Court (law) and Royal Colleges of Medicine and Surgery, became increasingly important as providers of professional training and regulation of competence starting in the Middle Ages. Over the centuries, governments promoted the teaching of 128 science, technology and advanced vocational training at universities, and occasionally even offered financial aid for their development, but nevertheless professional training remained mainly a private enterprise. It was not until the 19th century that major public universities were founded in England, Wales and Northern Ireland, which included the more applied disciplines into their fields of study (Smith and Langslow 1999).30 The foundation of university institutions in Germany started later than in Britain, during the 14th century. Erfurt was the first university to be founded in 1379 on what is today German territory (Verger 1993). The principle adopted was quite different to the established traditions, because it combined both types of university existing so far, the more professionally oriented and the more liberal one in a single higher education institution (Moraw 2005). For the next centuries, until the development of disciplines in the modern sense, four faculties prevailed at German universities: theology, law, medicine and the liberal arts. Due to the incorporation of two traditions, the aim of higher education was twofold: to educate administrative personnel for government and society, and to promote students’ personal development. Based on this dualism, the student population at German universities consisted of two separate groups, each of them with specific study preferences and labour market outcomes. Students born in nobility, on the one hand, predominantly focussed on the professionally-oriented subjects such as law or medicine and thereafter pursued careers in the upper levels of public administration or as doctors. Students from the lower or middle Bürgertum, in contrast, preferred to study liberal arts. After finishing university, they mainly returned to the region where they were born without having such explicit career prospects, many of them working in their fathers’ craft, as clergy for the Church, or as writers in regional administrations (Moraw 1993). These origins are still in place at German universities nowadays, which should have major implications for the labour market entry of higher education graduates. Comparing the university development in Britain and in Germany from its very beginning, history shows that both countries (or the regions that later became those countries) followed quite different paths regarding aims and objectives of higher education. While England adopted the ideal of a liberal education originating from the French tradition, Germany combined general 30 University education in Scotland also has a long, but very different history. Four universities – St Andrews, Glasgow, Aberdeen and Edinburgh, known collectively as the “four ancient Scottish universities” - were founded in the 15th and 16th centuries. From their beginnings, these higher education institutions concentrated from the beginning more on applied subjects than their English counterparts and stressed the importance of professional training, mainly to offer an alternative to the traditional universities of Oxford and Cambridge. Most of today’s Scottish higher education institutions tend to specialise in particular areas, for example in teacher education, arts and architecture, or health care, and in addition offer a broad range of vocationally orientated subjects at all levels. 129 and professional education by establishing the four faculty university. For centuries, professional training in England was not at all connected to university education, since its major function was to form the personality of the predominantly noble student body attending Oxford and Cambridge. And even though English universities adopted some of the professional principles, their focus on the liberal arts still remains an important component of undergraduate education today, especially at traditional universities. In Germany, in contrast, a dual-type university in terms of content and student body existed from the very beginning. Professional training always had constituted an integral part of German university education, even though Wilhelm von Humboldt explicitly tried to overcome the focus on applicable skills by advocating for the importance of general knowledge conferred through universities (Ellwein 1985; Humboldt 1956). Both traditions of university education, the liberal education as well as the integration of general and applied professional disciplines into one single university, continue to exist are still existent today in Germany and Britain, strongly influencing the importance of occupation-specific training by higher education institutions. 5.2.2 General Objectives of Higher Education Institutions These historical ideas associated with higher education are still in place today. German higher education places a much stronger emphasis on professional education and the applicability of knowledge, while British institutions still advocate the principle of a liberal education with a focus on more general skills, particular at the undergraduate level. In Germany, the importance of occupational specificity is indicated by the purpose of study at higher education institutions. According to the Framework Act for Higher Education, study at both universities and Fachhochschulen has to “prepare for occupational tasks that require the application of scientific and methodological knowledge or the ability of artistic expression” (BMBF 2002, §1, author's translation). In addition, the legislative framework requires “postsecondary education [to] prepare students for a particular field of occupation and to teach them the specific knowledge, skills and methods of the respective degree course, in order to enable them to perform scientific or artistic work and to act responsibly in a free, democratic and social state governed by the rule of law” (BMBF 2002, §1, author's translation).31 Thus, 31 „Sie [Die Hochschulen] bereiten auf berufliche Tätigkeiten vor, die die Anwendung wissenschaftlicher Erkenntnisse und wissenschaftlicher Methoden oder die Fähigkeit zu künstlerischer Gestaltung erfordern“ (BMBF 2002, § 1). „Lehre und Studium sollen die Studierenden auf ein berufliches Tätigkeitsfeld vorbereiten und ihnen die dafür erforderlichen fachlichen Kenntnisse, Fähigkeiten und Methoden dem jeweiligen Studiengang entsprechend so vermitteln, dass sie zu wissenschaftlicher oder künstlerischer Arbeit und zu 130 training for particular occupations or professions is an integral part of study at any German higher education institution, irrespective of types of degrees and fields of study. While this basic principle applies to both universities and Fachhochschulen, the former differ from the latter in that their education has traditionally been closely linked to basic and theoretical research and accordingly considered to teach abstract and general knowledge rather than applicable skills. But despite this emphasis on theory, there are several measures offered at universities to ensure that their students receive an occupation-specific training. First of all, proof of work experience (from four to six months, in some cases even up to a year) acquired before or while studying is required in a large variety of fields, especially in the natural and engineering sciences. To improve the employment prospects of humanities graduates whose field of study, being more general in nature, tend not to prepare students for specific occupations other than a career in academia, a lot of university institutions have set up programmes in collaboration with employment offices to equip these students with key skills during their studies (for example a grounding in computing or elementary business skills) and to place them in industry thereafter. Additionally, many universities offer programmes designed to prepare students for self-employment and to encourage them to set up their own businesses. In this regard, it is also helpful that the universities’ student counselling offices and the employment offices’ career guidance services furnish information and guidance to help graduates move from higher education into the labour market (Eurydice 2006). The importance of occupational specificity is even more pronounced at Fachhochschulen. Emphasis on the practical applicability of knowledge and specific reference to labour market requirements are characteristic features of the design of these degree courses and of the organisation of teaching and studying at this type of higher education institution. The purpose of occupation-specific training is served chiefly by incorporating one or two semesters of practical training within a business enterprise into every degree course offered (Praxissemester). Furthermore, it is quite common for students to derive their final thesis topic from problems they encountered during these training semesters. Some theses are even prepared in collaboration with industry and trade. All of these measures help students gain applied knowledge and insights into their field of specialisation and to establish contacts with prospective employers before graduating. The fact that teaching staff and course contents at Fachhochschulen are linked with applied research and development projects is also important for a high degree of occupation-specific training. Taken together, measures offered at verantwortlichem Handeln in einem freiheitlichen, demokratischen und sozialen Rechtsstaat befähigt werden“ (BMBF 2002, § 7). 131 universities and Fachhochschulen indicate a high importance of occupationspecific training provided at German higher education institutions. Though there are differences between universities and Fachhochschulen as to their extent and scope, their overall focus clearly points in the direction of the applicability of knowledge and close links to the labour market. As a consequence, German higher education can be characterised by a high degree of occupational specificity. Higher education students should face few problems in obtaining a job matching their qualification after graduation. Yet, intra-national variations in this regard might exist as well. On the one hand, Fachhochschulen degrees might lead to even closer matches between credentials and occupations as compared to university degrees. In contrast, more general fields of study such as the arts or the humanities should make it more difficult to find a matching job. In Britain, in contrast, the concept of an education that provides training in abstract thought and that values knowledge for its own sake has always formed part of the rationale behind higher education. For example, the Robbins Report (1963) specified four aims for higher education: instruction in skills suitable to play a part in the general division of labour; promoting the general powers of the mind; the advancement of learning; and the transmission of a common culture and common standards of citizenship. Only the first aim is directly geared towards labour market requirements, while all other three emphasise the general benefits of higher education, both for personal development and for participation in society. Interestingly, most White Papers and higher education acts over the past decades rarely mention the need to train British students for particular occupations. Rather, they are responding to general problems of mass higher education in relation to study organisation, funding of students and universities, or quality control. The most recent definition of British higher education was given by the 1988 Education Reform Act in Section 120, which describes higher education as ‘education provided by means of a course of any description mentioned in Schedule 6 of the Act’, that is, ‘a course of a standard higher than the standard of courses leading to General Certificate of Education Advanced-level (GCE A-level) or Business and Technology Education Council National Diploma or Certificate’. The courses listed include the ones already introduced in the section on institutional differentiation by type of degree, such as first degree courses, postgraduate courses (including higher degree courses, courses for the Diploma of Higher Education, courses for the Higher National Diploma (HND) or Higher National Certificate (HNC), courses for the Certificate in Education granting Qualified Teacher Status in Britain, or courses in preparation for a professional examination at a higherlevel. Only the last two explicitly refer to professional training, while the others are more general in nature. 132 The fact that liberal education and personal growth remain important principles in British higher education to this day can also be seen in the Dearing Report (1997) and in the Green Paper ‘The Learning Age’ (DfEE 1998), in which the British Government sets out its strategy for lifelong learning. The Dearing Report again stresses the intellectual and cultural purposes of higher education and states that higher education should aim to sustain a learning society through the intellectual development of individuals. Higher education should increase individuals’ knowledge and understanding both for their own sake and for the sake of the economy and society. The Government’s Green Paper (1998) argues in a similar manner, advocating for the major principle that anyone who has the capability for higher education should have the opportunity to benefit from it. Current objectives for higher education stress increasing and widening participation and improving standards and quality of teaching and learning (DfEE 1998). But despite the persistent emphasis on general knowledge and personal development rather than occupation-specific skills, labour market requirements have nevertheless gained more importance for British higher education. A first step was, as already mentioned, the introduction of polytechnics to advance industrial skills on a part- or full-time basis and to link higher education with industry and businesses (DES 1966). But since the 1980s in particular, occupational requirements have become more relevant for higher education policy making. The Government’s 1985 Green Paper on higher education stated that the design and content of courses should become more occupation-oriented by seeking advice from employers (DfE 1985). Most institutions responded to this advice by re-designing courses to take employers’ needs into account. More occupational courses have been introduced and the curriculum for more general courses has been adapted to cover key skills such as information and communication technology. Statistics showed that in 1994, occupation-specific training was the fastest growing area of British higher education (DfE 1994). The Dearing Report also recognised the part played by higher education in preparing students for work and recommended that institutions should seek to increase work experience for students (Dearing Report 1997). Today, sandwich courses are the main means for students to combine course work with work experience, but many higher education institutions now also offer opportunities for students to gain work experience as part of their degree. In addition, many employers regularly visit higher education institutions to give students the opportunity to discuss possibilities for employment. Much work is now on the way in this area, and funding has been provided by the Higher Education Quality and Employability Division of the DfEE for research and development projects to improve graduate employability. The latest reforms in particular, passed by the Labour Party elected in 1997, signalled a significant shift in the aims and functions of 133 higher education. In the 2003 White Paper and the resulting Higher Education Bill (2004), the Blair Government repeatedly stressed the point that higher education is to improve national economics by tightening the connection between higher education, employment, productivity and trade, and furthermore, to enhance student outcomes in employment-related skills and competencies (DfES 2003). Thus, the British development is two-folded. Even though legislation on higher education continues to emphasise the importance of more general skills and personal development, the necessity to respond to labour market needs has become more apparent, particularly during the 1990s. Nevertheless, the long historical tradition of liberal higher education is still in place. This should strongly influence the types of jobs obtained after graduation, leading to a lower match between higher education credentials and occupational outcomes. But intra-national differences might be in place as well, resulting in better matches of graduates with occupationspecific postgraduate degrees and degrees in more applied fields of study. The more detailed discussion of training for particular occupational groups will give further information in this regard. 5.2.3 Occupation-specific Training of Different Types of Degrees and Fields of Study The clear-cut differences between the importance of professional training in Germany and the focus on liberal education in Britain becomes even more apparent if particular types of degrees and fields of study are examined in more detail. The training of professionals, the training of engineers, and the training of managers serve as examples in this regard. This analysis again shows the tight connection between higher education and labour market requirements in Germany and the rather loose one in Britain. The Training of Professionals There is a longstanding sociological tradition to define a profession as an occupational group which, based on its claim to expert knowledge, enjoys a high degree of work autonomy, in return for professional self-regulation. But only recent years have seen a rising interest in the history of the professions in different countries. Cross-national research on professional development strikingly reveals that most of the established professions within a particular country developed in a similar way; however, national paths differ substantially. Especially Britain and Germany have had radically different historical trajectories of the evolution of professional systems (Abbott 1988). In Britain, the classical professions of law and medicine claimed their autonomy from the government as early as in the early 17th century and 134 thereby defined and circumscribed the role of the liberal state. In Germany, in contrast, autocratic rulers, both before and after national unification in 1871, incorporated the equivalent occupations into the state and defined the professional service as public service (Lane et al. 2000; Neal and Morgan 2000). As a consequence, the professions in Britain developed independently from the state, being the result of voluntary association. In Germany, in contrast, the state has played – and continues to play – an active role in the establishment, structuring and administration of the professions. The different historical trajectories have implications for the acquisition of professional qualifications. Since in Britain, the formation of independent professions preceded the secure establishment of higher education, professional autonomy has also meant control over education and training, accreditation, awarding of professional credentials, and admission to the professions (Lane et al. 2000). Even though academic routes were established in the course of development, only some of the professions, like those in medicine and teaching and recently, engineering, moved into the universities within the framework of a three to four year study programme (Lane 1989). Nowadays, universities in general provide a range of professionally accredited degree courses including engineering, accountancy, teacher training, librarianship and information science and medical studies. In particular some of the postgraduate degrees aim at teaching applicable knowledge for professional life (for example LLM for lawyers or MSc for engineers). However, qualifications specific to a profession and required for practice as a chartered member of the professional association still have to be obtained upon completion of a postgraduate degree, often through successfully completing further examination designed or accredited by professional bodies, such as the Chartered Institute of Public Finance and Accountancy and the Council of Legal Education. The ‘new’ universities, most of which were previously polytechnics, generally offer a wider range of courses leading to qualifications recognised by professional institutions. But despite these recent developments, it has been a characteristic of the British system of professions that professional associations remain to a large extent responsible for training future professionals. It is still true that no higher education qualification awards an automatic right to employment in a given profession. In Germany, in contrast, independent professions only arose when the university system was well established. Hence, professional education has been firmly in the hands of the state-regulated higher education system (Lane et al. 2000). In the beginning, this was particularly true for potential state servants such as lawyers, but during the early nineteenth century universities gradually expanded the professional subjects, introducing new academic degrees for areas considered desirable for the state. Until today, the purpose of studying at German higher education institutions is to prepare students for 135 a profession in a certain sphere of activity (BMBF 2002). Professional qualifications are conferred on the basis of examinations from universities and Fachhochschulen. At universities, professional training for the public sector is closely connected with a specific degree, the Staatsexamen as well as with particular fields of study. State examinations prepare students for professions of public importance, such as medicine, dentistry, veterinary medicine, pharmaceutics, food chemistry, law and the teaching profession (Eurydice 2006). An essential difference between state and other academic degrees taught at universities is the incorporation of practical training into courses with state examinations. The first state examination (Erstes Staatsexamen), which takes place after the theoretical part of studies, entitles students to work in the private sector only. Prospective public professionals, however, undergo an additional phase of practical training called preparatory service (Vorbereitungsdienst), which is concluded by a second state examination (Zweites Staatsexamen). Only this second examination grants them the right to practise their profession within the public service. Overall, the university in Germany and its close relationship to the state played a crucial role in the development of the German professions (MacDonald 1995). Professional associations, in contrast, neither have a direct influence on the content of training and accreditation nor on the number of new entrants into the profession. Overall, the link between the professions and higher education in Britain and Germany has not only differed in terms of prevalence, power, and influence of professional associations, but also in terms of their level of integration within the state. In Britain, professional associations to a large extent rationalise and control professional education and training. Even though today, the state-controlled higher education system offers professional training in some postgraduate degree courses, the licence to practice is still bound to further examinations by professional bodies. Institutionally, the university system therefore cannot be considered to be complementary to the system of professions, since occupation-specific training predominantly takes place outside, previous examples notwithstanding. In Germany, it is the government that prescribes and sanctions the kind of training required for entry into established or new professions. Higher education is thus complementary to the systems of professions, since higher education degrees form the necessary prerequisite for entrance. A further interesting perspective in this regard is offered by Hage (2000) in his analysis of institutional path dependencies of elite education in Germany and Britain. By looking at the historical nature of elite education, he argues that countryspecific features influence the professional division of labour by means of the relative emphasis on generalist vs. specialist education. In this respect German and British higher education are both specialised, but in quite different ways. While the Germans emphasise scientific and technical 142 and professional labour markets in Germany and Britain are largely the result of different historical trajectories in the development of professions in both countries, which will be explored in the following. At the same time, these national forms of professional regulation have important consequences for the organisation of professional labour markets. The following sections will pay more attention to the relationship between professions and the state and the resulting degree of social closure for professional labour markets. The Historical Relation between Professional Governance and the State The major difference between German and British professions is related to the importance of state regulation. German professions are largely stateregulated, and the individual professional chambers are subject to statutory supervision. In contrast, British professions were for a long time largely selfregulated by professional bodies, with hardly any interference by the Government (Abbott 1988; Lane et al. 2000; Neal and Morgan 2000). Both trajectories are a legacy of history, according to which German professionalisation has been the result of state intervention, while development in Britain has been more spontaneous and driven by professional self-interest. In Germany, the non-existence of an overarching civil service during the 19th century laid the foundation for the establishment of professions (Lane et al. 2000; Neal and Morgan 2000). Germany was not united until 1871 and even then consisted of 25 individual states, each having different political and cultural characteristics. German rulers of these states had wide-ranging authority and powers as compared to the federal government. They were at this time supported by a well trained but authoritarian civil service, which was largely the result of Napoleon’s reform in Germany closely mirroring the French administrative structure. These involved reforms of many key institutions in the German states, such as the school system, the church, the legal system, and the establishment of expert ministries supported by a professional civil service. As a result, bureaucratic authorities and professions developed hand in hand in Germany. Autocratic rulers, both before and after 1871, incorporated professions such as law, medicine, or teaching, into the state and defined professional service as public service. At the same time, the lack of a central legislative authority was a serious deterrent to the development of fully independent associations, as were common in the Anglo-Saxon world at that time. National professional organisations could not evolve until the late 19th century, and even then, strong regional differences were maintained. In Britain, the development of the British Empire and the beginning of the Industrial Revolution led to the rapid expansion of a wealthy middle class and early demand for all types of professional expertise in the 17th century 143 (Lane et al. 2000; Neal and Morgan 2000). Early professions such as legal advice in setting up commercial enterprises, accountancy, or auditing work, started to organise and professionalise at the national level without state interference. The laissez-faire policies of successive British governments in the 19th century further enabled and encouraged the emerging occupations to organise themselves in order to fend off competitors. For self-interested reasons, these groups were highly motivated to achieve social closure: to gain self-determination, and to establish barriers against to entry by related occupations. In direct contrast to Germany, social status did not come from state recognition, but from being a member of the relatively independent professional classes. Establishing professional associations was seen as way for members to follow an established route to success, sheltered from free competition, which would exert a downward pressure on prices and services (Neal and Morgan 2000). This legacy of the past substantially influenced the development of professions in both countries and ensured their social closure. This was done by different means, however. Even today, the government regulates professional activity in Germany through laws, while in Britain regulation is rarely issued from parliament, and self-regulation through professional associations is more common. The most important difference in this regard is again related to professional training, since German legal regulation secured monopolies of occupational knowledge and practice at higher education institutions, while the British Parliament only protects the occupational title and usually has no control over occupational training. In this regard, the previous sections have already shown that in Germany the government continues to be largely responsible for professional education and admittance to the professions. Entry to high status professional positions is usually gained by means of higher education credentials. In Britain, in contrast, professional associations remain responsible for professional education, even though in the course of development some academic routes have been established. Thus, particular types of higher education credentials should be more important for German entry-ports to the respective profession than British ones. This should be strongly the case for the Staatsexamen, but also for specific fields of study such as health or education. At the same time, entry into the professions should occur much earlier after graduation in Germany than in Britain, since no additional entry certificates have to be obtained from professional associations. The close link between higher education and the professions in Germany and the dominance of professional training outside higher education by professional associations in Britain can be considered a direct consequence of the different historical trajectories. However, the traditional British stateprofessional relationship has been radically transformed since the early 1980s. Margaret Thatcher introduced reforms that served to eliminate some 144 of the self-regulatory advantage that British professions previously enjoyed, since the British Government has come to doubt that self-governing professions will act in the public interest. Hence, a succession of Acts from 1984 onwards have introduced the dual mechanisms of increasing market competition and more external supervision into the British system of professions, which strongly weakened the social closure of professional labour markets. The different boundaries established by professions in both countries to keep out competitors and the latest developments in Britain in this regard will be described in the following section. The Social Closure of Professional Labour Markets In both countries, markets for professional services have traditionally been highly protected, providing professions with either a market shelter or a cartel-like arrangement and more rarely with a genuine monopoly. As a result of recent reforms in Britain, however, the principle of uncontrolled professional self-governance has increasingly been questioned since the 1980s by the introduction of market elements such as increasing competition, more external quality assurance and cost containment (Lane et al. 2000). As a consequence, attitudes towards market regulation and what is deemed a legitimate suspension of competition have become more diverse between the British and German system. Professional labour markets in Germany have always been sheltered almost completely from competition by various legal arrangements (Lane et al. 2000). By and large, suspension of competition has been much more prevalent, more effectively enforced and more legitimate in Germany than in Britain. The many regulations in restraint of competition regarding the activities of professions are still in place and not in danger of being revoked. For example, German lawyers still enjoy a legally enshrined virtual monopoly in all legal matters, laid down by the Law on Legal Advice and Legal Activity. At the same time, methods of calculating fees for professional services are still largely regulated by law. Only recently, external competition has started to emerge, when law practices opened up to related professions with similar constitutions, such as accountants. Furthermore, intraprofessional competition has become more severe in the labour market of lawyers through steadily rising numbers of graduates moving in. But even in the course of higher education expansion in Germany, unemployment among new law graduates is not substantively higher than among professionals in general, and absolute figures are low (Lane et al. 2004). Overall, German professions still enjoy a high degree of market shelter closely representing occupational labour market characteristics. The only exception to this is the trend that fixed-term employment has risen considerably among German professionals, particularly in the law profession, but also in academia. But 145 despite this development, career development of German professionals should closely follow an internal labour market pattern. Entry into this segment of the labour market should be strongly based on higher education credentials. Career mobility should largely proceed within the professional domain, and little mobility out of the segment should occur. Although the British professions have been exempted from some aspects of the Competition Act (1998), they have nevertheless have become more and more exposed to a high degree of inter- and intra-professional competition. Governmental efforts to secure cost containment have also put limits on professionals’ endeavour to determine their own levels of rewards. While these measures of increased external supervision and direct intervention began under the Thatcher government, Labour has continued this interventionist policy. These moves towards state-circumscribed selfgovernment in combination with attempted marketisation and cost-cutting of publicly financed service in British professions have strongly reduced the social closure of the professional labour market traditionally secured by professional bodies. As a consequence, British professionals have experienced a dramatic increase in potential and actual competition from other professions and non-professionalised occupations. It is even argued that the traditional boundaries between various professional services are crumbling and will completely disappear (Lane 2000). Graduate career mobility in Britain should be less predetermined by the professional labour market segment. Due to the weak linkage between higher education and professions it should it be less common to directly enter the professions immediately after graduation. But even if entry has been successful, movements out of this segment should be more pronounced as a result of increasing marketisation. In summary, the tight coupling between the state and the professions in Germany, and the self-regulation and increasing market competition in the British system of professions can be considered as empirical indicators of the fact that specific forms of labour market segmentation also prevail in both countries regarding the organisation of professions. Regarding entry into the professions, one can conclude that the link between the professions and higher education in Britain and Germany is different not only in terms of prevalence, power, and influence of professional associations, but also in terms of their level of integration with the state: While in Britain, professional associations rationalise and control professional education and training, in Germany it is the government that prescribes and sanctions the kind of training required for entry into established or new professions. Abbott (1988) refers in this regard to Britain as a country where a strong state was largely absent during the creation of modern professions. By contrast, the German government created the institutions that trained professionals in order to provide the loyal civil servants that are needed. The 146 previous sections have shown how public service has historically played a different role in both countries in shaping the development of professions. The importance of further differences of public sector organisation for graduate career mobility will be discussed in the following section. 5.3.2 Public Services between Status Protection and New Public Management Since the “golden age” of the welfare state, the public sector has developed into one of the largest sectors. In some countries, it has even become the largest employer throughout Europe. At the same time, the public sector has, to a large extent, been an important labour market for higher education graduates. However, the European public sector is characterised by a strong degree of diversity, since the definition of public service employment, its career structure and qualification requirements vary strongly across countries. The following section gives an overview of entry requirements in the public sector, examines career prospects in this labour market segment, and analyses the size and development of public sector employment in Germany and Britain. Special emphasis is given to reforms of the public sector during the 1980s and 1990s. To start with, it is important to note that the definition of public employment varies strongly between the two countries, which makes comparisons rather complicated. Country-specific definitions of civil servants are especially diverse. In Germany, civil servants are called Staatsbeamte and civil service is regulated by the Beamtenrecht, the law of public service. The work of the British ‘civil servant’ is regulated not by law, but by the Civil Service Code. The size of these groups also varies strongly between the two countries: while German civil servants account for almost half of all public employees, their British counterparts constitute only a small minority (Rothenbacher 1999; 2001/2002). In order to overcome the obstacle of comparing societal groups that are not fully comparable, this study adopts a rather broad perspective by analysing the development of employment within the public sector in general and not of civil servants in particular, since comparative data are more often available. Only for a better understanding of employment opportunities in both countries the example of civil servants is employed in order to make cross-national diversity more explicit. Qualification Requirements for Entry into the Public Service In both countries, the rise of the nation state had profound consequences for the patterns of control and administration of public service and its relation to 147 the university system. In Germany, forging the nation state went hand in hand with incorporating academia into the ranks of public service, thereby placing upon it the implicit obligation of service to the national community. British public service institutions have historically been less integrated with universities; instead they have tended to formulate their own entry requirements for civil servants (Neave 2001). As a consequence of this development, direct ties between German universities and public service emerged in form of the dual status of higher education qualifications – the split between academic qualifications (Diplom, Magister), which did not qualify for public appointments, and state qualifications (Staatsexmanen), which entitled the holder to apply for employment in public administration. In Britain, public service reforms and higher education reforms were kept in two separate spheres, and the status of the university as a self-governing institution remained intact. If the British civil service indorsed the principle of appointment on the basis of competitive public examinations, such examinations were not imposed on the universities. In Germany, the dual employment structure of public service is part of this historical tradition. The core elements of public service are relatively uniform for public servants at all levels of government, with the term “public servant” being used as a generic term to include civil servants (Beamte) as well as public employees (Angestellte). Due to these traditions, posts for civil servants have been established for law and order functions and must be reserved for them – according to article 33, paragraph 4, Basic Law – while all other public service tasks are mainly fulfilled by public employees. Even though the dual employment structure made sense in the last century, it has become less and less appropriate and relevant, since the boundaries between the two categories of service law have become increasingly blurred in practice (Becker, R. 1993). In keeping with the basic principles of a merit-based career system, entrance for civil servants into the civil service classes is strictly linked to certain formal qualification requirements (Becker, R. 1993; Kuhlmann and Röber 2004). The German civil servants’ professional profile is largely shaped by the existing system of education and recruitment which combines professional training and career structure. For entering the upper level of German civil service, the so called ”administrative class”, a university degree is required which usually has to be completed by a two years preparatory service (“Referendariat”) with practical stages in different public and private institutions and some courses on administrative topics. But there is no special education or training for the administrative class. A condition to enter the intermediate level of the civil service, the ”executive class”, is – after having passed the Abitur-examination – a degree at a college for public administration. Similar qualification requirements, however, apply to public 148 employees as well, with higher hierarchical positions being reserved for higher education graduates. This again points to the fact that despite the continued dual employment structure, the two categories of public service employment have in practice been coming closer to each other. An important aspect of training for the German public service is that no central elite schools for the education and training exist. In contrast to the still centralised French system of civil service training offered by the Ecole Nationale d’Administation (ENA), in Germany all higher education institutions are in principle equally responsible for education and training of public servants, even though the universities of Potsdam and Konstanz and the German School for Administrative Sciences in Speyer offer special courses in administrative sciences. Not surprisingly, a central unit responsible for recruiting civil servants or public employees does not exist. Each ministry at the federal level, each state, and each local authority has the right to recruit its staff members itself (Becker, R. 1993; Kuhlmann and Röber 2004). In Britain, such a tight coupling between higher education credentials and the public service has traditionally not existed. The intake of higher education graduates into public employment only started to emerge with increasing higher education expansion. Higher education graduates comprised only five percent of recruits at the executive officer level in 1965, but around half the appointments at this level by the 1980s. However, particularly at the upper levels of the civil service hierarchy, charges of elitism have always been made. In this regard, Oxford and Cambridge have always been central institutions for the formation of an administrative elite. Whilst only 15 percent of those applying for high-ranked posts are from Oxbridge, they make up for over half of the total who enter (Dowding 1995). Apart from this, however, British higher education never established a separate curricular pathway for public service employees within the undergraduate syllabus. Rather, holding a university degree was deemed a prior condition for eligibility to sit for the public examination, results of which decided who should be called and who should be rejected. In this regard, competition continues to be an important feature of the public management bureaucracy. In particular, it emphasises competition in the form of a keen rivalry among the “best and the brightest” of university graduates for appointment to “fast stream” places in civil service (Hood, Christopher 2003). National differences can also be found regarding the fields of study important for entry into public service in both countries. According to the ”legalistic” administrative culture which has been generated by the prevalence of legal rule-application in Germany’s administrative business, the personnel structure in public administration is very strongly moulded by a dominance of lawyers (Kuhlmann and Röber 2004). The proportion of 149 lawyers in Federal ministries, amounting to 63%, has virtually remained unchanged for more than 30 years, succeeded by barely 15% of economists (Derlien 2003). In Britain, more general fields of study have been more important for entry into the public service. Far greater numbers of humanities and arts graduates apply to the civil service, with history being a traditionally highly valued subject based on its high degree of general education. There is also a strong bias towards arts degree holders amongst permanent secretaries, while specialist subjects have always offered fewer career opportunities than the generalist grades. The lower career opportunities for specialists in the British public service has always been justified on the grounds that specialists are too deeply involved in technical matters to take a detached view (Dowding 1995). More similarities exist as regards subject requirements for professions in the public service which have traditionally been highly specialised in both countries. But national differences persist as well concerning the status of individual professional groups in the public sector. One of the biggest groups of public service employees in both countries are teachers. However, while German primary and secondary school teachers as well as higher education professors have the status of a civil servant, they do not belong to the civil service in Britain. Overall, higher education in Germany has always been much more tightly connected with entry into the public service than in Britain. Even though higher education there also constitutes a necessary precondition, further examinations to select candidates from the best graduates are still in place. Consequently, entry to the administrative class, be it civil servants of public employees, should be much easier for German higher education graduates immediately after graduation. Career Structure in the Public Service There are two basic models for employment in the core public service, a career-based and a position-based system . The choice of one system or the other has a profound effect on the career trajectories of public service employees. The first one closely represents the model of firm-internal labour markets, while the second exhibits more external labour market structures. In career-based systems, public servants are expected to stay in the public service more or less throughout their working life. Initial entry is based on academic credentials and/or a civil service entry examination. Once recruited, people are placed in positions by the organisation. This may include moving staff from one ministry to another or from one area of specialisation to another. Promotion is based on a system of grades attached to the individual rather than to a specific position. This sort of system is characterised by limited possibilities for entering the civil service mid-career and a strong emphasis on career development. Compared to the private sector public 150 sector employees are rewarded for long-time employment with career stability. Position-based systems focus on selecting the best-suited candidate (OECD 2004c) for each position, whether by external recruitment or internal promotion. Thus, they are more similar to the private sector, allow for more open access, and lateral entry is relatively common. Traditionally, both Germany and Britain exhibited strong characteristics of a career-based system. However, while this is still the case in Germany nowadays, Britain has moved from a strongly career-based system to a strongly position-based system in the past two decades following the reforms of its public sector (OECD 2004c). The essential feature of German civil service in its present form goes back to the end of the eighteenth century in the Prussian General Code of 1794. The prevailing philosophy had been confirmed in the “Reich Civil Service Act” (Reichsbeamtengesetz) enacted in 1873 (Siedentopf 1990: 236) and was enshrined after the Second World War in the German constitution especially by ”reserving to civil servants the right to act on behalf of the state” (article 33, paragraph 4, Basic Law) and by emphasising the traditional principles of professional civil service (article 33, paragraph 5, Basic Law) (Siedentopf 1990: 237, cited in Kuhlmann and Röber 2004). Although there is no clearly defined enumeration of elements which constitute the traditional principles of the civil service, some features are widely seen as characteristic, e.g. lifetime occupation, an appropriate salary according to the maintenance principle (Alimentationsprinzip), loyalty, political neutrality and moderation, dedication to public service, no right to strike and subjection to special disciplinary regulations. However, traditional principles of professional civil service only apply to Beamte and are not aimed at regulating the legal status of Angestellte, who are subject to private sector law and public sector industrial relations. But despite these legal differences, career prospects are more similar, since the principle of seniority also applies to public employees, with pay-scales and promotion strongly prompting a continuous career development with the same employer. German public service has been very resistant to reform. During the 1980s, the lack of new policy initiatives in West Germany, despite the triumph in 1983 of the Conservative Coalition government, was striking, especially compared to Britain. Yet, the new government did not change the basic policy approach to public sector organisation. Attempts at privatisation and de-regulation were almost non-existent before being implemented on a comparatively modest scale towards the end of the 1990s. Even the new challenges of the late 1980s, namely the impact of unification as well as the adaptive pressures that result from the ongoing process of European integration did not lead to far-reaching adaptations of the German public sector. Though the localities were successful in reducing some of their load of responsibilities by cut-back, retrenchment and contracting-out, 151 privatisation and de-regulation initiatives of the public service have, until relatively recently, been rare occurrences despite a political rhetoric which suggested otherwise (Hesse 2003). Only recently, the reform overload following from these developments is documented by the growing awareness of the necessity to reduce the number of public servants and public employees significantly. German public service is not likely to escape scrutiny. However, the system of checks and balances – entrenched as much in German political culture as it is solidified in institutional relations – neither required nor encouraged dramatic structural changes as observed in Britain (Hesse 2003; Röber and Löffler 2000). In Britain, the regulation of civil service employment dates back to the Northcote-Trevelyan Report of 1854. This Report established the British civil service in its modern form, with a division between superior and inferior posts corresponding to intellectual and mechanical tasks, merit-based entry upon competitive examinations, training-on-the-job, as broad a range of subjects as possible, and promotion based on merit (Dowding 1995). Until the late 1970s, competition for promotion on a closed career basis among an elite corps once appointed was a marked feature of the traditional British civil service, because it put the onus on providing information about loyalty and good service to the organisation on those best placed to provide it (i.e. employees) rather than review procedures or output-based pay systems (Hood, Christopher 2003). Until then, employment relations across the British public sector were distinct from those found in private firms in a number of key respects. Perhaps the most important difference was the high level of centralisation of employment relations policy. For most groups, pay determination was (and largely remains) linked to national structures of collective bargaining or, in the case of some professions (for example, the medical profession), pay review bodies. These structures offered little scope for local variation or interpretation of standardised grading structures and terms and conditions (Horton 2000). Also, the role of the state as employer served to standardise practice, demonstrated by the focus on continuity of employment, procedural fairness in collective bargaining, investment in training and equal opportunities. This status quo remained rather unchallenged until Margaret Thatcher came into power in 1979, which ultimately introduced market dynamics into public service. Thatcher introduced market features into the so far shielded public sector. Much of the thrust of the reforms now dubbed as “New Public Management” (NPM) during the 1980s and 1990s was ostensibly designed to add new forms of competition to the traditional forms of public sector competition (in the form of individual rivalry for appointment, honours and promotion, and organisational competition for budgetary funds and policy responsibility), to release managerial energies within the public sector by relaxing rules that stifled managerial creativity, and put public sector 254 Brennan, J.; Lyon, S.; Schomburg, H.; Teichler, U. (1996b). Employment and Work of British and Germany Graduates. In J. Brennan, M. Kogan & U. Teichler (Eds.), Higher Education and Work. London, Bristol, Pennsylvania: Jessica Kingsley Publishers, pp. 47-98. Briedis, K.; Minks, K.-H. (2004). Zwischen Hochschule und Arbeitsmarkt. Eine Befragung der Hochschulabsolventinnen und Hochschulabsolventen des Prüfungsjahres 2001. Hannover: HIS Hochschul-Informations-System. Büchel, F.; De Grip, A.; Mertens, A. (2003). Overeducation in Europe: Current Issues in Theory and Policy. Cheltenham, UK: Edward Elgar. Butz, M. (2000). Lohnt Bildung noch? Die Einkommensungleichheiten der unterschiedlichen Bildungsklassen in der Bundesrepublik Deutschland im Zeitvergleich 1982-1995. In P. A. Berger & D. Konietzka (Eds.), Neue Ungleichheiten der Erwerbsgesellschaft. Opladen: Leske & Budrich, pp. 95-117. Bynner, J.; Butler, N.; Ferri, E.; Shepherd, P.; Smith, K. (2001). The Design and Conduct of the 1999-2000 Surveys of the National Child Development Study and the 1970 British Cohort Study, CLS Cohort Studies Working Papers. London: Centre for Longitudinal Studies. Chevalier, A.; Conlon, G. (2003). Does it Pay to Attend a Prestigious University? Centre for the Economics of Education, London School of Economics and Political Science, London. Clark, B. R. (1995). Places of Inquiry. Research and Advanced Education in Modern Universities. Berkeley, CA: University of California Press. Clasen, J. (2005). Reforming European Welfare States. Germany and the United Kingdom Compared. Oxford, New York: Oxford University Press. Coleman, J. S. (1994). Foundations of Social Theory. Cambridge MA., London: The Belknap Press of Harvard University Press. Collins, R. (1979). The Credential Society. New York: Academic Press. CRE (2000). The Bologna Declaration on the European Space for Higher Education: An Explanation. Paris: Confederation of EU Rectors’ Conferences and the Association of European Universities. Davies, S.; Guppy, N. (1997). Field of Study, College Selectivity, and Student Inequalities in Higher Education. Social Forces, 75, pp. 1417-1438. Deakin, S.; Reed, H. (2003). River Crossing or Cold Bath? Deregulation and Employment in Britain in the 1980s and 1990s. In G. Esping-Andersen & M. Regini (Eds.), Why Deregulate Labour Markets? Oxfort, New York: Oxford University Press. Dearing Report (1997). Higher Education in the Learning Society, Report submitted to the Secretaries of State for Education and Employment, Wales, Scotland 255 and Northern Ireland in July 1997. http://www.leeds.ac.uk/educol/ncihe/: The National Committee of Inquiry into Higher Education. Dekker, R.; de Grip, A.; Heijke, H. (2002). The Effects of Training and Overeducation on Career Mobility in a Segmented Labour Market. International Journal of Manpower, 23, pp. 106-125. Derlien, H.-U. (2003). German Public Administration: Weberian despite "Modernization". In K. K. Tummala (Ed.), Comparative Bureaucratic Systems. Lanham, Boulder, New York, Oxford: Lexington Books, pp. 97- 122. DES (1966). A Plan for Polytechnics and Other Colleges. Higher Education in the Further Education System (Cm. 3006), Great Britain. Parliament. House of Commons. London: HMSO. DES (1991). Higher Education: A New Framework. Cm. Paper/Bill Number: 1541, Great Britain. Parliament. House of Commons. London: HMSO. DfE (1985). The Development of Higher Education into the 1990s (Cm. 9524), Great Britain. Parliament. House of Commons. London: HMSO. DfE (1994). Student Numbers in Higher Education - Great Britain 1982/83 to 1992/93, Great Britain. Parliament. House of Commons. London: HMSO. DfEE (1998). Green Paper: The Learning Age: A Renaissance for a New Britain, Great Britain. Parliament. House of Commons. London: HMSO. DfES (2003). The Future of Higher Education. Government White Paper presented to Parliament by the Secretary of State for Education and Skills by Command of Her Majesty, Great Britain. Parliament. House of Commons. London: HMSO. Doeringer, P. B. (1967). Determinants of the Structure of Industrial Type Internal Labour Markets. Industrial Relations Review, 20, pp. 206-220. Doeringer, P. B.; Piore, M. (1971). Internal Labor Markets and Manpower Analysis. Lexington (MA): Heath. Dolton, P.; McIntosh, S. (2003). Public and Private Sector Labour Markets. In R. Dickens, P. Gregg & J. Wadsworth (Eds.), The Labour Market under New Labour: The State of Working Britain. Houndmills, Basingstoke, New York: Palgrave Macmillan, pp. 214-232. Dowding, K. (1995). The Civil Service. London, New York: Routledge. Ebbinghaus, B. (2005). Can Path Dependence Explain Institutional Change?, Max Planck Institute for Study of Societies Discussion Papers. Berlin: MPIfG. Ebbinghaus, B.; Manow, P. (2001). Introduction. Studying Varieties of Welfare Capitalism. In B. Ebbinghaus & P. Manow (Eds.), Comparing Welfare Capitalism. Social Policy and Political Economy in Europe, Japan and the USA. London: Routledge, pp. 1-24. 256 Elder, G. H. (1985). Perspectives on the Life Course. In G. H. Elder (Ed.), Life Course Dynamics. Trajectories and Transitions, 1968-1980. Ithaca: Cornell University Press, pp. 23-49. Elder, G. H. (1992). The Life Course. In E. Borgatta & M. Borgatta (Eds.), The Encyclopedia of Sociology, Volume 3. Houndmills, Basingstoke, New York: Palgrave Macmillan, pp. 1120-1130. Elder, G. H. (1994). Time, Human Agency, and the Social Change: Perspectives on the Life Course. Social Psychology Quarterly, 57, pp. 4-15. Elder, G. H.; Johnson Kirkpatrick, M.; Crosnoe, R. (2003). The Emergence and Development of Life Course Theory. In J. T. Mortimer & M. J. Shanahan (Eds.), Handbook of the Life Course. New York: Kluwer Academic Publisher, pp. 3-19. Elias , P. (1997). Occupational Classification (ISCO-88): Concepts, Methods, Reliability, Validity and Cross-national Comparability. OECD Labour Market and Social Policy Occasional Papers, 20, pp. 1-23. Elias, P.; Purcell, K. (2003). Measuring Change in the Graduate Labour Market. Project Paper 'Researching Graduate Careers Seven Years On', 1. Ellwein, T. (1985). Die deutsche Universität. Vom Mittelalter bis zur Gegenwart. Königstein: Athenäum Verlag. Enders, J. (2004). Higher Education, Internationalisation, and the Nation-State: Recent Developments and Challenges to Governance Theory. Higher Education, 47, pp. 361-382. Erikson, R.; Goldthorpe, J. H. (1992). The Constant Flux. A Study of Class Mobility in Industrial Societies. Oxford: Clarendon Press. Erikson, R.; H., G. J.; Portocarero, L. (1979). Intergenerational Class Mobility in Three Western European Societies. British Journal of Sociology, 30, pp. 415-441. Estévez-Abe, M.; Iversen, T.; Soskice, D. (2001). Social Protection and the Formation of Skills: A Reinterpretation of the Welfare State. In P. A. Hall & D. Soskice (Eds.), Varieties of Capitalism. The Institutional Foundations of Comparative Advantage. Oxford, New York: Oxford University Press, pp. 145-183. European Commission (2004). Employment in Europe 2004. Recent Trends and Prospects. Luxembourg: Office for Official Publications of the European Communities. Eurydice (2000). Two Decades of Reform in Higher Education in Europe: 1980 Onwards. Brussels: Eurydice, the Information Network on Education in Europe. 257 Eurydice (2006). Structures of Education, Vocational Training and Adult Education Systems in Europe - 2006 Edition. Brussels: Eurydice, the Information Network on Education in Europe. Falk, S.; Sackmann, R.; Struck, O.; Weymann, A.; Windzio, M.; Wingens, M. (2000). Gemeinsame Startbedingungen in Ost und West? Risiken beim Berufseinstieg und deren Folgen im weiteren Erwerbsverlauf Working Paper Series. Bremen: Collaborative Research Center "Status Passages and Risks in the Life Course" (Sfb 186). Fuchs, S.; Schettkat, R. (2003). Germany: A Regulated Flexibility. In G. Esping- Andersen & M. Regini (Eds.), Why Deregulated Labour Markets? Oxford, New York: Oxford University Press. Gangl, M. (2000a). Education and Labour Market Entry across Europe: The Impact of Institutional Arrangements in Training Systems and Labour Markets, Working Papers Mannheimer Zentrum für Sozialforschung. Mannheim: MZES. Gangl, M. (2000b). European Perspectives on Labour Market Entry: A Matter of Institutional Linkages and Labour Markets?, Working Papers Mannheimer Zentrum für Sozialforschung. Mannheim: MZES. Gangl, M. (2001). European Perspectives on Labour Market Entry: A Dichotomy of Occupationalized versus Non-Occupationalized Systems? European Societies, 3, pp. 471-494. Gangl, M. (2002a). Changing Labour Markets and Early Career Outcomes: Labour Market Entry in Europe over the Past Decade. Work, Employment and Society, 16, pp. 67-90. Gangl, M. (2002b). The Only Way is Up? Employment Protection and Job Mobility among Recent Entrants to European Labour Markets, Working Papers Mannheimer Zentrum für Sozialforschung. Mannheim: MZES. Gangl, M. (2004). Institutions and the Structure of Labour Market Matching in the United States and West Germany. European Sociological Review, 20, pp. 171-187. Grimshaw, D.; Ward, K. G.; Rubery, J.; Beynon, H. (2001). Organisations and the Transformation of the Internal Labour Market. Work, Employment and Society, 15, pp. 25-54. Hage, J. (2000). Path Dependencies of Education Systems and the Division of Labour within Organisations. In M. Maurice & A. Sorge (Eds.), Embedding Organisations. Societal Analysis of Actors, Organisations and Socio- Economic Context. Amsterdam, Philadelphia: John Benjamins Publishing Company. Haisken-DeNew, J. P.; Frick, J. R. (2005). DTC Desktop Companion to the German Socio-Economic Panel Study (SOEP). Berlin: DIW. 258 Hall, P. A.; Soskice, D. (2001). An Introduction to Varieties of Capitalism. In P. A. Hall & D. Soskice (Eds.), Varieties of Capitalism. The Institutional Foundations of Comparative Advantage. Oxford: University Press, pp. 1- 68. Hammouya, M. (1999). Statistics on Public Sector Employment: Methodology, Structures and Trends, ILO Bureau of Statistics Working Paper. Geneva: International Labour Office. Hannan, D. F.; Raffe, D.; Rutjes, H.; Willems, E.; Mansuy, M.; Müller, W.; Amor, T. (1999). A Comparative Analysis of Transitions from Education to Work in Europe (CATEWE) - A Conceptual Framework, ESRI Working Paper Dublin: European Studies Research Insitute. Hartmann, J. (1997). Komplexes Stichprobendesign und Ereignisanalyse. Zur Notwendigkeit einer Gewichtung bei disproprtional geschichteter Stichprobenziehung, Working Papers Mannheimer Zentrum für Sozialforschung. Mannheim: MZES. Heckman, J. J. (2002). Flexibility and Job Creation: Lessons for Germany, National Bureau of Economic Research Working Paper Series. Cambridge (MA): NBER. Heidenreich, M. (1999). Berufskonstruktion und Professionalisierung. Erträge der soziologischen Forschung. In H.-J. Apel, K.-P. Horn, P. Lundgreen & U. Sandfuchs (Eds.), Professionalisierung pädagogischer Berufe im historischen Prozess. Bad Heilbrunn: Julius Klinkhardt, pp. 35-58. Heinz, W. R. (1996). Status Passages as Micro-Macro Linkages in Life Course Research. In A. Weymann & W. R. Heinz (Eds.), Society and Biography. Weinheim: DSV, pp. 51-66. Heinz, W. R. (1999). From Education to Work: Cross-National Perspectives. Cambridge: Cambridge University Press. Hesse, J. J. (2003). Stability Turned Regidity. Paradoxes in German Public Sector Reform. In J. J. Hesse, C. Hood & B. G. Peters (Eds.), Paradoxes in Public Sector Reform. An International Comparison. Berlin: Duncker & Humblot, pp. 197-214. Hill, D. H. (1997). Adjusting for Attrition in Event-History Analysis. Sociological Methodology, 27, pp. 393-416. Hillmert, S. (2001). Ausbildungssysteme und Arbeitsmarkt : Lebensverläufe in Großbritannien und Deutschland im Kohortenvergleich. Wiesbaden: Westdeutscher Verlag. Hillyard, P.; Sim, J.; Tombs, S.; Whyte, D. (2003). Leaving a "Stain Upon the Silence": Critical Ciminology and the Politics of Dissent, "Tough on Crime" ... Though on Freedoms? From Community to Global Interventions. Centre for Studies in Crime and Social Justice (Edge Hill), Chester, England. 259 Hood, C. (1995). The "New Public Management" in the 1980s: Variations on a Theme. Accounting, Organizations and Society, 20, pp. 93-109. Hood, C. (2003). From Public Bureaucracy State to Re-regulated Public Service: The Paradox of British Public Sector Reform. In J. J. Hesse, C. Hood & B. G. Peters (Eds.), Paradoxes in Public Sector Reform. An International Comparison. Berlin: Duncker & Humblot, pp. 127-148. Horton, S. (2000). Human Resources Flexibilities in UK Public Services. In D. Farnham & S. Horton (Eds.), Human Resources Flexibilities in the Public Services. Houndmills, Basingstoke, New York: Palgrave Macmillan. Huisman, J.; Maassen, P.; Neave, G. (2001). Higher Education and the Nation State. The International Dimension on Higher Education. Oxford: Pergamon. Humboldt, W. v. (1956). Über die innere und äußere Organisation der höheren wissenschaftlichen Anstalten in Berlin (1810). In n.a. (Ed.), Die Idee der deutschen Universität: Die fünf Grundschriften aus der Zeit ihrer Neubegründung durch klassischen Idealismus und romantischen Idealismus. Darmstadt: Wissenschaftliche Buchgesellschaft. Hussmanns, R.; Mehran, F.; Verma, V. (1990). Surveys of Economically Active Population, Employment, Unemployment and Underemployment: An ILO Manual on Concepts and Methods. Geneva: International Labour Organization. Jahr, V.; Schomburg, H.; Teichler, U. (2003). Internationale Mobilität von Absolventinnen und Absolventen europäischer Hochschulen. Kassel: Wissenschaftliches Zentrum für Berufs- und Hochschulforschung. Jenkins, S. P. (2004). Survival Analysis, Manuscript of the Essex Summer School in SSCA. University of Essex, Colchester. Jongbloed, B. (2004). Funding Higher Education: Options, Trade-Offs and Dilemmas. Fulbright Brainstorms - New Trends in Higher Education. CHEPS, University of Twente, the Netherlands. Jovanovic, B. (1979). Job Matching and the Theory of Turnover. The Journal of Political Economy, 5, pp. 972-990. Kalleberg, A. L.; Sorensen, A. B. (1979). The Sociology of Labor Markets. Annual Review of Sociology, 5, pp. 351-379. Keller, A. (2003). Von Bologna nach Berlin. Perspektiven eines Europäischen Hochschulraums im Rahmen des Bologna-Prozesses am Vorabend des europäischen Hochschulgipfels 2003 in Berlin. Berlin. Kerckhoff, A. C. (1996). Building Conceptual and Empirical Bridges between Studies of Educational and Labour Force Careers. In A. C. Kerckhoff (Ed.), Generating Social Stratification: Toward a New Research Agenda. Boulder: Westview Press, pp. 37-56. 260 Kerckhoff, A. C. (2001). Education and Social Stratification. Processes in Comparative Perspective. Sociology of Education, Extra Issue 2001, pp. 3- 18. Kerr, C. (1954). The Balkanization of Labor Markets. In E. W. Bakke, P. M. Hauser, G. L. Palmer, C. A. Myers, D. Yoder & C. Kerr (Eds.), Labor Mobility and Economic Opportunity Cambridge (MA): Technology-Press of MIT, pp. 92-110. Kerst, C.; Minks, K.-H. (2003). Fünf Jahre nach dem Studienabschluss - Berufsverlauf und aktuelle Situation von Hochschulabsolventinnen und Hochschulabsolventen des Prüfungsjahrgangs 1997. In H. Hochschul- Informations-System (Ed.), HIS Projektbericht. Hannover: HIS. Kim, A.; Kim, K.-W. (2003). Returns to Tertiary Education in Germany and the UK: Effects of Fields of Study and Gender, Working Papers Mannheimer Zentrum für Sozialforschung. Mannheim: MZES. Kim, A.; Kurz, K. (2001). Precarious Employment, Education and Gender: A Comparison of Germany and the United Kingdom, Working Papers Mannheimer Zentrum für Sozialforschung. Mannheim: MZES. Kivinen, O.; Nurmi, J. (2003). Unifying Higher Education for Different Kinds of Europeans. Higher Education and Work: A Comparison of Ten Countries. Comparative Education, 39, pp. 83-103. Klemm, K. (2000). Bildung. In J. Allmendinger & W. Ludwig-Mayerhofer (Eds.), Soziologie des Sozialstaates. Weinheim: Juventa, pp. 145-166. Kogan, I.; Müller, W. (2003). School-to-Work Transitions in Europe: Analyses of the EU LFS 2000 Hoc Module. Introduction. In I. Kogan & W. Müller (Eds.), School-to-Work Transitions in Europe: Analyses of the EU LFS 2000 Hoc Module. Mannheim: MZES. Kohli, M. (1985). Die Institutionalisierung des Lebenslaufs: Historische Befunde und theoretische Argumente. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 37, pp. 1-29. Kohlrausch, B. (2007). A Ticket to Work? Active Labour Market Policy for the Young Unemployed in Britain and Germany. Dissertation, University of Bremen. Kohlrausch, B. (2009). A Ticket to Work? Policies for the Young Unemployed in Britain and Germany. Frankfurt/M., New York Campus Verlag. Korsnes, O. (2000). Towards a Relational Approach to the Study of the Variety in Situated Creativity of Economic Actors. In M. Maurice & A. Sorge (Eds.), Embedding Organizations. Societal Analysis of Actors, Organisations and Socio-economic Context. Amsterdam, Philadelphia: John Benjamins Publishing Company, pp. 71-88. Krais, B. (1996). Bildungsexpansion und soziale Ungleichheit in der Bundesrepublik Deutschland. In A. Bolder, W. R. Heinz & K. Rodex (Eds.), Die 261 Wiederentdeckung der Ungleichheit. Jahrbuch '96 Bildung und Arbeit. Opladen: Leske & Budrich, pp. 118-146. Kuhlmann, S.; Röber, M. (2004). Civil Service in Germany: Characteristics of Public Employment and Modernization of Public Personnel Management, Modernization of State and Administration in Europe: A France-Germany Comparison. Bordeaux, France: Goethe-Institut. Lane, C. (1989). Management and Labour in Europe: The Industrial Enterprise in Germany, Britain and France. Aldershot: Edward Elgar Publishing Limited. Lane, C.; Potton, M.; Littek, W. (2000). The Professions between State and Market. A Cross-national Study of Convergence and Divergence. ESRC Working Paper. Lane, C.; Wilkinson, F.; Littek, W.; Heisig, U.; Brown, J.; Burchell, B.; Mankelow, R.; Potton, M.; Tutschner, R. (2004). The Future of Professionalised Work in Britain and Germany. Solicitors and Advocates. In Anglo-German Foundation for the Study of Industrial Society (Ed.), Anglo-German Foundation Reports. London. Leff, G. (1993). Die Artes Liberales. In W. Rüegg (Ed.), Geschichte der Universität in Europa. Band 1. München: Beck. Leisering, L.; Leibfried, S. (1999). Time and Poverty in Western Welfare States. Cambridge: Cambridge University Press. Leuze, K. (2007). Sex or Subject? What Makes the Difference for Sex Segregation in Graduate Labour Markets?, 3rd International SOEP Young Scholars Symposium. Hanse Institute for Advanced Studies, Delmenhorst, Germany. Li, Y. (2002). Falling off the Ladder? Professional and Managerial Career Trajectories and Unemployment Experiences. European Sociological Review, 18, pp. 253-270. Lindberg, M. E. (2007). Connections between the Differentiation of Higher Education Participation and the Distribution of Occupational Status. A Comparative Study of Seven European Countries. European Societies, 9, pp. 551-572. Lutz, B.; Sengenberger, W. (1974). Arbeitsmarktsegmentation und öffentliche Beschäftigungspolitik. Göttingen: Otto Schwarz. Lutz, B.; Sengenberger, W. (1980). Segmentationsanalyse und Beschäftigungspolitik. WSI-Mitteilungen, 5, pp. 291-299. MacDonald, K. (1995). The Sociology of the Professions. Beverly Hills, London: Sage Publications. March, J. G.; Olson, J. P. (1989). Rediscovering Institutions. New York: The Free Press. Marsden, D. (1990). Institutions and Labour Mobility: Occupational and Internal Labour Markets in Britain, France, Italy and West Germany. In R. Brunetta 262 & C. Dell'Aringa (Eds.), Labour Relations and Economic Performance. Houndmills, Basingstoke, New York: Palgrave Macmillan, pp. 414-438. Marsden, D. (1999). A Theory of Employment Systems: Micro-Foundations of Societal Diversity. Oxford: Oxford University Press. Martens, K.; Rusconi, A.; Leuze, K. (2007). New Arenas of Education Governance. The Impact of International Organizations and Markets on Educational Policymaking. Houndmills, Basingstoke, New York: Palgrave Macmillan. Mason, G. (2000). Production Supervisors in Britain, Germany and the United States: Back from the Dead Again? Work, Employment and Society, 14, pp. 625- 645. Maurice, M. (2000). The Paradox of Societal Analysis. A Review of the Past and Prospects for the Future. In M. Maurice & A. Sorge (Eds.), Embedding Organisations. Societal Analysis of Actors, Organisations and Socioeconomic Context. Amsterdam, Philadelphia: John Benjamins Publishing Company, pp. 13-37. Maurice, M.; Sellier, F.; Silvestre, J.-J. (1986). The Social Foundations of Industrial Power: A Comparison of France and Germany. Cambridge, MA: MIT Press. Mayer, K. U. (2001). The Paradox of Global Social Change and Institutional Path Dependencies. In A. Woodward & M. Kohli (Eds.), Inclusions and Exclusions in European Societies. London, New York: Routlegde, pp. 89- 110. Mayer, K. U.; Müller, W. (1989). Lebensverläufe im Wohlfahrtsstaat. In A. Weymann (Ed.), Handlungsspielräume. Untersuchungen zur Individualisierung und Institutionalisierung von Lebensläufen in der Moderne. Stuttgart: Enke, pp. 41-61. Meyer, J. W. (1977). The Effects of Education as an Institution. American Journal of Sociology, 83, pp. 55-77. Meyer, J. W.; Ramirez, F. O.; Soysal, Y. N. (1992). World Expansion of Mass Education, 1870-1980. Sociology of Education, 65, pp. 128-149. Meyer, J. W.; Rowan, B. (1977). Institutionalized Organizations: Formal Structures as Myth and Ceremony. American Journal of Sociology, 83, pp. 340-363. Mincer, J. (1970). The Distribution of Labor Incomes: A Survey With Special Reference to the Human Capital Approach. Journal of Economic Literature, 8, pp. 1-26. Moraw, P. (1993). Der Lebensweg der Studenten. In W. Rüegg (Ed.), Geschichte der Universität in Europa. Band 1. München: Beck. Moraw, P. (2005). Die Universitäten in Europa und Deutschland. Anfänge und Schritte auf einem langen Weg. In U. Sieg & D. Korsch (Eds.), Die Idee der Universität heute. München: K.G. Saur, pp. 25-41. 263 Müller, W. (1998). Erwartete und unerwartete Folgen der Bildungsexpansion. In J. Friedrichs, M. R. Lepsius & K. U. Mayer (Eds.), Die Diagnosefähigkeit der Soziologie, Sonderheft 38 der Kölner Zeitschrift für Soziologie. Opladen: Leske & Budrich, pp. 81-112. Müller, W. (2000). Education and Labour Market Outcomes: Commonality or Divergence? In M. Haller (Ed.), The Making of the European Union. Contributions of the Social Sciences. Berlin, Heidelberg, New York: Springer-Verlag, pp. 287-308. Müller, W.; Bruns, H.; Steinemann, S. (2002). Expansion und Erträge tertiärer Bildung in Deutschland, Frankreich und im Vereinigten Königreich. Berliner Journal für Soziologie, 12, pp. 37-62. Müller, W.; Gangl, M. (2003). Transitions from Education to Work in Europe. The Integration of Youth into EU Labour Markets. Oxford, New York: Oxford University Press. Müller, W.; Shavit, Y. (1998). The Institutional Embeddedness of the Stratification Process. A Comparative Study of Qualifications and Occupations in Thirteen Countries. In Y. Shavit & W. Müller (Eds.), From School to Work. Oxford: Clarenden Press, pp. 1-48. Neal, M.; Morgan, J. (2000). The Professionalization of Everyone? A Comparative Study of the Development of the Professions in the United Kingdom and Germany. European Sociological Review, 16, pp. 9-26. Neave, G. (2001). The European Dimension of Higher Education. In J. Huisman, P. Maassen & G. Neave (Eds.), Higher Education and the Nation State. The International Dimension on Higher Education. Oxford: Pergamon. North, D. C. (1990). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. OECD (2002). Highlights of Public Sector Pay and Employment Trends: 2002 Update. Paris: OECD Publishing. OECD (2003). Learning for Tomorrow's World - First Results from PISA 2003. Paris: OECD Publishing. OECD (2004a). Education at a Glance. OECD Indicators 2004. Paris: OECD Publishing. OECD (2004b). OECD Handbook for Internationally Comparative Education Statistics: Concepts, Standards, Definitions and Classifications. Paris: OECD Publishing. OECD (2004c). Public Sector Modernisation: Modernising Public Employment. In OECD (Ed.), Policy Brief July 2004. Paris: OECD Publishing. Pierson, P. (2001). Post-Industrial Pressures on the Mature Welfare States. In P. Pierson (Ed.), The New Politics of the Welfare State. Oxford, New York: Oxford University Press, pp. 80-104. 270 Appendix B: The Erikson-Goldthorpe Class Scheme Classes Occupations included I Higher-grade professionals, administrators, and officials; managers in large industrial establishments; large proprietors II Lower grade professionals, administrators, and officials; higher-grade technicians; managers in small industrial establishments; supervisors of non-manual employees IIIa Routine non-manual employees, higher grade (in administration and commerce) IIIb Routine non-manual employees, lower grade (sales and services) IVab Small proprietors and artisans with or without employees IVc Farmers and smallholders; other self-employed in primary production V Lower-grade technicians; supervisors of manual workers VI Skilled manual workers VIIa Semi- and unskilled manual workers (not in agriculture) VIIb Agricultural and other workers in primary production (Source: Erikson, Robert and Goldthorpe 1992: 38f) Appendix C: The New Casmin Educational Classification Qualification Description 1a Inadequately completed general education 1b General elementary education 1c Basic vocational qualification or general elementary education and vocational qualification 2a Intermediate vocational qualification or intermediate general qualification and vocational qualification 2b Intermediate general qualification 2c_gen General maturity certificate 2c_voc Vocational maturity certificate/General maturity certificate and vocational qualification 3a Lower tertiary education 3a_gen Lower tertiary education – general diplomas 3b_voc Lower tertiary education – diplomas with vocational emphasis 3b Higher tertiary education 3b_low Higher tertiary education – lower level 3b_high Higher tertiary education – higher level [Document text truncated for crawler view.]